Comparability of Cause of Death Between ICD-9

Volume 49, Number 2 May 18, 2001 Comparability of Cause of Death Between ICD–9 and ICD–10: Preliminary Estimates by Robert N. Anderson, Ph.D.; Arialdi M. Miniño, M.P.H.; Donna L. Hoyert, Ph.D.; and Harry M. Rosenberg, Ph.D.

Abstract Introduction Objectives—This report presents preliminary results describing This report presents preliminary data describing the effects of the effects of implementing the Tenth Revision of the International the implementation of the Tenth Revision of the International Classi- Classification of Diseases (ICD–10) on mortality statistics for selected fication of Diseases (ICD–10) on mortality statistics for selected causes of death effective with deaths occurring in the United States causes of death. ICD–10 was implemented in the United States in 1999. The report also describes major features of the Tenth beginning with deaths occurring in 1999 and replaces the Ninth Revision (ICD–10), including changes from the Ninth Revision Revision of the ICD (ICD–9), which was implemented in the United (ICD–9) in classification and rules for selecting underlying causes of States with 1979 mortality data. death. Application of comparability ratios is also discussed. The International Classification of Diseases has been revised Methods—The report is based on cause-of-death information from approximately every 10 years since 1900 (1) (table A). The purpose of a large sample of 1996 death certificates filed in the 50 States and the the revision is to stay abreast of medical advances in terms of disease District of Columbia. Cause-of-death information in the sample includes nomenclature and etiology. The introduction of new classifications is underlying cause of death classified by both ICD–9 and ICD–10. costly to the Federal government and the States, and often introduces Because the data file on which comparability information is derived is major disruptions in time series of mortality and morbidity statistics. incomplete, results are preliminary. However, revisions are essential to stay current with advances in Results—Preliminary comparability ratios by cause of death pre- medical science and to ensure the international comparability of health sented in this report indicate the extent of discontinuities in cause-of- statistics. death trends from 1998 through 1999 resulting from implementing ICD–10 differs from ICD–9 in several respects. ICD–10 is far more ICD–10. For some leading causes (e.g., Septicemia, Influenza and detailed than ICD–9, with about 8,000 categories compared with about pneumonia, Alzheimer’s disease, and Nephritis, nephrotic syndrome 5,000 categories; ICD–10 uses alphanumeric codes compared with and nephrosis), the discontinuity in trend is substantial. The ranking of numeric codes in ICD–9; some additions and modifications were made leading causes of death is also substantially affected for some causes to the chapters in the ICD; and some of the coding rules and rules for of death. selecting the underlying cause of death have been changed. Measures Conclusions—Results of this study, although preliminary, are of the discontinuities in cause-of-death statistics resulting from these essential to analyzing trends in mortality between ICD–9 and ICD–10. classification and rule changes are critical to the interpretation of In particular, the results provide a means for interpreting changes mortality trends. between 1998, which is the last year in which ICD–9 was used, and 1999, the year in which ICD–10 was implemented for mortality in the Acknowledgments United States. This report was prepared in the Mortality Statistics Branch, Division of Vital Keywords: comparability c mortality c ICD–10 c bridge-coding Statistics. The Registration Methods staff and the Data Acquisition and Evaluation Branch provided consultation to State vital statistics offices regarding the collection of the death certificate data on which this report is based. Nicholas F. Pace, Chief of the Systems Programming and Statistical Resources Branch, provided computing support in the creation of the comparability file. Donna E. Glenn, June S. Pearce, Julia E. Raynor, and Joyce L. Bius of the Mortality Medical Classification Branch, provided technical support and nosological review of changes in classification and rules for selecting underlying cause of death. Wayne E. Johnson and Van L. Parsons of the Office of Research and Methodology provided statistical expertise and contributed to the Technical notes. 2 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001

Table A. Implementation dates of ten revisions of the (ICD–9 codes 401.0, 401.1, 401.9), for example, has been reduced to International Classification of Diseases a single code in ICD–10 (I10—with no 4th digit subcategories). ICD–10 uses a 4-digit alphanumeric coding scheme compared Years in use Revision by U.S. with the 4-digit numeric codes used in ICD–9. Each of the 21 chapters of ICD–10 is classified to a letter or letters of the alphabet (see table B). First...... 1900–09 Infectious disease codes in chapter I, for example, begin with an ‘‘A’’ Second...... 1910–20 Third ...... 1921–29 or ‘‘B.’’ Thus, Acute poliomyelitis is associated with the codes Fourth ...... 1930–38 A80.0–A80.9 and Viral hepatitis is classified as B15.0–B19.9. The Fifth...... 1939–48 4-digit alphanumeric scheme used with ICD–10 allows a larger number Sixth ...... 1949–57 Seventh ...... 1958–67 of codes than the 4-digit numeric scheme used with ICD–9. Eighth, adapted ...... 1968–78 Chapters of the ICD have been added and rearranged. Chapter Ninth ...... 1979–98 titles for ICD–9 and ICD–10 are shown in table B. ICD–9 chapter VI, Tenth ...... 1999– Diseases of the nervous system and sense organs, was divided into three chapters in ICD–10: chapter VI, Diseases of the nervous system; chapter VII, Diseases of the eye and adnexa; and chapter VIII, Dis- Studies of the comparability between revisions of the ICD have eases of the ear and mastoid process. Also, External causes of been carried out and published in the United States at least since the morbidity and mortality and Factors influencing health status and Fifth Revision (2–5). Comparability studies—also called bridge-coding contact with health services, which were supplementary classifications studies—are designed to measure the effects of a new revision of the in ICD–9, have been assigned as chapters XX and XXI, respectively, ICD on the comparability with the previous revision of mortality statistics in ICD–10. Chapters III (Endocrine, nutritional and metabolic diseases by cause of death. These studies involve the dual classification of a and immunity disorders) and IV (Diseases of the blood and blood- single year’s mortality data, i.e., classifying the underlying cause of forming organs) in ICD–9 are exchanged in ICD–10. Chapter IV now death on mortality records by both the new revision and the previous contains the endocrine, nutritional and metabolic diseases, and chapter revision. The key element of a comparability study is the comparability III contains the diseases of blood and blood-forming organs. Immune ratio, which is derived from the dual classification. It is calculated by disorders remain in chapter III. Diseases of the genitourinary system dividing the number of deaths classified by the new revision by the (previously chapter X) and Complications of pregnancy, childbirth, and number of deaths classified by the previous revision. The resulting the puerperium (previously chapter XI) have become chapters XIV and ratios represent the net effect of the new revision on cause-of-death XV, respectively, in ICD–10. statistics and can be used with prudence as factors to adjust mortality Cause-of-death titles have been changed and regrouped. statistics for causes of death classified by a previous revision to be Examples of title changes include the ICD–9 title Chronic obstructive comparable to rates for the same causes classified by the new revision. pulmonary diseases and allied conditions, which became Chronic lower This report discusses major features of the Tenth Revision and respiratory diseases in ICD–10. ICD–9’s Suicide became Intentional changes in classification and rules for selecting underlying causes of self-harm, and Homicide became Assault in ICD–10. Notable regroup- death from the Ninth Revision. Comparability ratios based on prelimi- ings include some cerebrovascular disorders, specifically transient nary data are presented that describe the net effects of the Tenth cerebral ischemic attacks, which have been moved from Diseases of Revision on mortality statistics for selected causes of death. Application the circulatory system (ICD–9 code 435) to Diseases of the nervous of comparability ratios is also discussed. system (ICD–10 codes G45.8 and G45.9). Septic shock, classified in ICD–9 as Shock without mention of trauma (785.5) in Symptoms, signs, and ill-defined conditions, is classified in ICD–10 to Unspecified septicemia (A41.9) in chapter I (Certain infectious and parasitic diseases). Major features of ICD–10 and changes in Respiratory failure (799.1) was moved from Symptoms, signs, and classification ill-defined conditions to Diseases of the respiratory system (J96). Myelodysplastic syndromes were moved from Diseases of the blood Several new features and changes from ICD–9toICD–10 will and blood-forming organs (289.8) to Neoplasms of uncertain behavior affect the classification of cause of death. ICD–10 is much more (D46). End stage renal disease, classified under Other disorders of detailed with about 8,000 categories for classifying causes of death kidney and ureter in ICD–9 (593.9), has been reclassified in ICD–10 compared with nearly 5,000 in ICD–9. Generally, with successive as Renal failure (N18.0). Transport accidents have been regrouped by revisions, the trend is to expand disease categories to provide more the characteristics of the injured person (e.g., pedestrian, pedal cyclist, detailed information about the type or site of the disease. For motorcycle rider, car occupant). In ICD–9, transport accidents were example, the single code for acute myocardial infarction (ICD–9 code grouped by the type of vehicle involved in the accident. 410—with no 4th digit subcategories) is expanded in ICD–10 to six different codes (I21.0—I21.4, I21.9), which specify the site of the infarction. Also, an increased number of terms have been assigned Changes in international selection and perinatal codes. For example, while ICD–9 provides a single code for modification rules uremia (585) regardless of the age of the decedent, ICD–10 provides a perinatal code for uremia. Thus, in ICD–10, for decedents less than Mortality statistics by cause of death are compiled from entries 28 days of age, uremia is coded P96.0 and N19 for all other ages. In on the medical certification portion of the death certificate that follows a few cases, less detail is shown. Essential hypertension the World Health Organization (WHO) format (6). Causes of death National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 3

Table B. Chapter titles for the Ninth and Tenth Revisions, International Classiﬁcation of Diseases Chapter ICD–91 chapter titles (code range2) ICD–101 chapter titles (code range2)

I Infectious and parasitic diseases ...... (001–139) Certain infectious and parasitic diseases ...... (A00–B99) II Neoplasms ...... (140–239) Neoplasms ...... (C00–D48) III Endocrine, nutritional and metabolic diseases and Diseases of the blood and blood-forming organs and certain immunity disorders ...... (240–279) disorders involving the immune mechanism ...... (D50–D89) IV Diseases of blood and blood-forming organs ...... (280–289) Endocrine, nutritional and metabolic diseases ...... (E00–E90) V Mental disorders ...... (290–319) Mental and behavioral disorders ...... (F00–F99) VI Diseases of the nervous system and sense organs ...... (320–389) Diseases of the nervous system ...... (G00–G99) VII Diseases of the circulatory system ...... (390–459) Diseases of the eye and adnexa ...... (H00–H59) VIII Diseases of the respiratory system ...... (460–519) Diseases of the ear and mastoid process ...... (H60–H95) IX Diseases of the digestive system ...... (520–579) Diseases of the circulatory system ...... (I00–I99) X Diseases of the genitourinary system ...... (580–629) Diseases of the respiratory system ...... (J00–J99) XI Complications of pregnancy, childbirth and the puerperium .....(630–676) Diseases of the digestive system ...... (K00–K93) XII Diseases of the skin and subcutaneous tissue ...... (680–709) Diseases of the skin and subcutaneous tissue ...... (L00–L99) XIII Diseases of the musculoskeletal system and connective tissue . . . (710–739) Diseases of the musculoskeletal system and connective tissue . . (M00–M99) XIV Congenital anomalies ...... (740–759) Diseases of the genitourinary system ...... (N00–N99) XV Certain conditions originating in the perinatal period ...... (760–779) Pregnancy, childbirth and the puerperium ...... (O00–O99) XVI Symptoms, signs and ill-defined conditions ...... (780–799) Certain conditions originating in the perinatal period ...... (P00–P96) XVII Injury and poisoning ...... (800–999) Congenital malformations, deformations and chromosomal abnormalities ...... (Q00–Q99) XVIII - - - Symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified...... (R00–R99) XIX - - - Injury, poisoning and certain other consequences of external causes ...... (S00–T98) XX - - - External causes of morbidity and mortality ...... (V01–Y98) XXI - - - Factors influencing health status and contact with health services ...... (Z00–Z99) - - - Supplementary classification of external causes of injury and poisoning ...... (E800–E999) ------Supplementary classification of factors influencing health status and contact with health services...... (V01–V82) - - -

- - - Category not applicable. 1ICD–9isInternational Classification of Diseases, Ninth Revision, and ICD–10 is International Classification of Diseases, Tenth Revision. 2The fourth digits of the upper and lower limits of the code ranges are not shown. include ‘‘all those diseases, morbid conditions or injuries which either Certificate (7,8). The U.S. Standard Certificate, in turn, closely follows resulted in or contributed to death and the circumstances of the the format recommended by the WHO. accident or violence which produced these injuries’’ (6: p. 30). The If the death certificate is properly completed, the disease or medical certification of death is divided into two sections (see condition listed on the lowest line used in Part I is usually accepted as figure 1). In Part I, the physician is asked to provide the causal chain the underlying cause of death. This is an application of ‘‘The General of morbid conditions that led to death, beginning with the condition Principle’’ (6). The General Principle is applied unless it is highly most proximate to death on line (a) and working backwards to the improbable that the condition on the lowest line used could have given initiating condition. The lines (a) through (d) in Part I are connected rise to all of the diseases or conditions listed above it. In some cases, by the phrase ‘‘due to, or as a consequence of.’’ They were designed the sequence of morbid events entered on the death certificate is not to encourage the physician to provide the causally related sequence specified correctly. A variety of errors may occur in completing the of medical conditions that resulted in death. Thus, the condition on medical certification of death. Common problems include the following: line (a) should be due to the condition on line (b), and the condition the causal chain may be listed in reverse order; the distinction between on line (b) should be a consequence of the condition on line (c), etc., Part I and Part II may have been ignored so that the causal sequence until the full sequence is described back to the originating or initiating in Part I is simply extended unbroken into Part II; or the reported condition. If only one step in the chain of morbid events is recorded, underlying cause is unlikely, in an etiological sense, to have caused a single entry on line (a) is adequate. Part I of the medical the condition listed above it. In addition, sometimes the physician certification is designed to facilitate the selection of the underlying attributes the death to uninformative causes such as cardiac arrest or cause of death when two or more causes are recorded on the pulmonary arrest. certificate. The underlying cause of death is defined as ‘‘(a) the To resolve the problems of incorrect or implausible cause- of-death disease or injury which initiated the train of morbid events leading statements, the WHO designed standardized rules to select an under- directly to death, or (b) the circumstances of the accident or violence lying cause of death from the information available on the death which produced the fatal injury’’ (6: p. 31) and is generally considered certificate that is most informative from a public health perspective. The the most useful cause from a public health standpoint. Part II solicits rules for the Tenth Revision as updated by WHO since publication of other conditions that the physician believed contributed to death, but ICD–10 (6) are described in an NCHS instruction manual (8). The rules were not in the causal chain. Figure 1 shows the medical certification for the Ninth Revision are described in an earlier ICD–9 edition of the part of the U.S. Standard Certificate of Death. While some details of same NCHS instruction manual and in Volume I of the ICD–9 manual the death certificate vary by State, all States use the same general (9). Coding rules beyond the ‘‘General Principle’’ are invoked if the format for medical certification outlined in the U.S. Standard cause-of-death section is completed incorrectly or if their application 4 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001

Figure 1. Medical certification portion of the U.S. Standard Certificate of Death can improve the specificity and characterization of the cause of death to make neoplasms secondary or primary. Thus, if cancer of the in a manner consistent with the ICD. The rules are applied in two steps: stomach is listed in Part I, line (a), and cancer of the breast is listed selection of a tentative underlying cause of death, and modification of in line (b), both are considered primary sites because neither is listed the tentative underlying cause in view of the other conditions reported as a common site of metastasis. In ICD–9, cancer of the breast would on the certificate in either Part I or Part II. Modification involves several be selected as the primary and underlying cause of death because it considerations by the medical coder: determining whether conditions is listed on the lowest line used in Part I. In ICD–10, when two or more in Part II could have given rise to the underlying cause, giving pref- malignant neoplasms are considered primary, the underlying cause of erence to specific terms over generalized terms, and creating linkages death is assigned to Malignant neoplasms of independent (primary) of conditions that are consistent with the terminology of the ICD. multiple sites (C97) except when the sites are in the same organ The rules for selecting the underlying cause of death have been system. In this case, the underlying cause is coded to the ill-defined reevaluated and in some instances changed. Several changes to the site of the organ system. For example, if stomach cancer is listed in Ninth Revision rules and the interpretation of these rules have been Part I, line (a), and cancer of the gallbladder is listed in line (b), the made with the Tenth Revision. (See table I for a summary comparison underlying cause should be Ill-defined sites within the digestive system of ICD–9 and ICD–10 rules for selecting the underlying cause of death.) (C26.9). In ICD–9, the underlying cause would be selected on the basis Some of the more prominent changes are discussed below. Other of order of entry, i.e., cancer of the gallbladder. changes and their effects on specific causes of death are discussed In ICD–10, secondary cancers are acceptable as the underlying in the Results section. cause of death if no primary site is specified. Thus, if metastatic brain In general, the international selection and modification rules for carcinoma is the only condition certified, it is selected as the underlying ICD–10 are similar to ICD–9. In structure, two rules were consolidated cause of death in ICD–10. In ICD–9, secondary cancers were invalid (i.e., those involving senility and ill-defined conditions) and two rules as an underlying cause of death. In the aforementioned example where were dropped (i.e., those involving old pneumonia, influenza, and only metastatic brain carcinoma is listed, the ICD–9 underlying cause maternal conditions, and errors and accidents in medical care). The would be Other malignant neoplasm without specification of site greatest impact on statistical data reflects changes in details under (199.1). rules and coding instructions. Most prominent is a change in the rules In ICD–10, lung has been added to the list of common sites of for direct sequels (Rule 3). The ICD–10 direct sequel rule states, ‘‘[i]f metastasis and is considered secondary whenever it appears in Part the condition selected by the General Principle or by Rule 1 or Rule I with any other site not on the list. Thus, if cancer of colon is listed 2 is obviously a direct consequence of another reported condition, in line (a) and cancer of lung in line (b), then cancer of lung is whether in Part I or Part II [of the medical certification portion of the considered secondary and cancer of colon is selected as the underlying death certificate], select this primary condition’’ (6: p. 34). The cause cause. However, if cancer of lung were to be listed in line (a) and cancer of death most affected by Rule 3 is pneumonia, which is often the of brain in line (b), cancer of lung would be selected as the underlying consequence of another condition or injury. In ICD–10, the applicability cause of death because brain is also listed as a common site of of Rule 3 to pneumonia is broader than in ICD–9 so pneumonia is metastasis. Although these changes do not have an impact on the considered a consequence of a much wider range of conditions. As a malignant neoplasm category as a whole, they do have an impact on result, pneumonia is much less likely to be selected as the underlying the distribution of underlying cause of death among various subcat- cause of death under ICD–10 than under ICD–9. It follows that several egories of malignant neoplasms. other conditions, including a small number of heart diseases, Two important changes have been made to the rules that regulate cerebrovascular diseases, complications of diabetes, malignant neo- how senility and ill-defined conditions are handled (Rule A). First, plasms, and other wasting diseases, when listed with pneumonia, will although Sudden infant death syndrome (SIDS) is classified in the be more likely to be selected. ill-defined conditions chapter in ICD–10 (chapter XVIII), SIDS is exempt Also important are several changes in the way the primary site of from Rule A. Rule A states specifically that ‘‘[w]here the selected cause malignant neoplasms is selected. In ICD–10, order of entry is not used is classifiable to chapter XVIII. . . except for R95 [SIDS]. . . and a National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 5 condition classified elsewhere...isreported on the certificate, reselect cause under ICD–9. For example, the preliminary comparability ratio the cause of death as if the condition classified to chapter XVIII had for Viral hepatitis is 0.8343 (table 1). In ICD–9, 1,346 deaths were not been reported, except to take account of that condition if it modifies classified as Viral hepatitis. In ICD–10, when the same deaths were the coding’’ (6: p. 42). In ICD–9, SIDS is classified as ill-defined and classified, the number of deaths due to Viral hepatitis fell to 1,123, a is subject to the ill-defined rule (ICD–9, Rule 5). Thus, SIDS is likely decrease of 223 deaths or about 16.5 percent. This means that Viral to be selected as the underlying cause of death more often under hepatitis was less likely to be selected as the underlying cause of death ICD–10. Second, several conditions not contained in chapter XVIII are in ICD–10 than in ICD–9. treated as ill-defined in ICD–10 (7). These include unspecified cardiac Comparability ratios greater than 1.00 are the result of a larger arrest (I46.9), unspecified hypotension (I95.9), Other and unspecified number of deaths being classified to cause i under ICD–10. For disorders of circulatory system (I99), Acute respiratory failure (J96.0), example, the comparability ratio for Septicemia is 1.1949 (table 1). In Unspecified respiratory failure (J96.9), and Respiratory failure of new- ICD–9, 17,791 deaths were classified to Septicemia. In ICD–10, 21,258 born (P28.5). Under the ICD–9 Rule 5, these causes of death are not deaths were classified to Septicemia, an increase of 3,467 deaths or treated as ill-defined conditions. nearly 20 percent. Thus, Septicemia is more likely to be selected as Changes in the instructions regarding highly improbable the underlying cause of death in ICD–10 than in ICD–9. sequences were made with ICD–10. Each of the following was considered an improbable sequence under ICD–9 rules and an acceptable sequence in ICD–10: a) Certain infectious and parasitic diseases can Data and methods be due to certain diseases outside of chapter I; b) any infectious disease can be due to disorders of the immune mechanism, chemo- Data therapy, and tumors of the immune system; c) any malignant neoplasm The data in this report are based on cause-of-death information can be due to human immunodeficiency virus (HIV) infection. Under from a large nonrandom sample of 1996 death certificates filed in the ICD–9, only a few specified malignant neoplasms could be due to HIV. 50 States and the District of Columbia. The cause-of-death informa- d) Diabetes can be considered due to malnutrition; e) Intracranial tion in the sample includes underlying cause of death classified by hemorrhage can be due to certain heart diseases (I05–I08, I09.1, both ICD–9 and ICD–10. The sample comprises 1,852,671 out of the I33–I38) and diseases of the digestive system; and f) Cerebral embo- total 2,314,690 (80 percent) resident deaths occurring in the United lism can be due to diseases of the digestive system. States during 1996. All records that could be coded in time for the preparation of this report were included. Because the data file on Comparability ratios which comparability information is derived is incomplete, comparability results are preliminary. A description of the sample is included The classification and rule changes between ICD–10 and ICD–9 below. have a considerable impact on some of the major causes of death in the United States. The shifting of deaths away from some cause-of- death categories and into others resulting from these changes Classification of underlying cause creates discontinuities in cause-of-death trends from 1998, the last The underlying cause of death for each record in the 1996 year of ICD–9, and 1999, the first year of ICD–10. The comparability mortality data was originally classified by ICD–9 (10). A major task of ratios presented in this report are designed to measure these this study was to reclassify 1996 mortality records by ICD–10 to discontinuities and are key to understanding the trend in mortality provide an underlying cause of death classified by each revision. This statistics from 1998 to 1999. was done primarily by using the ICD–10 versions of the mortality The comparability ratios presented in this report are based on medical software programs designed to select the underlying cause coding the same deaths occurring in 1996 by both the Ninth and Tenth of death from electronic death record data. Two pieces of this suite of Revisions and measure the net effect of ICD–10 by cause of death. software are particularly involved. The Mortality Medical Indexing, Operationally, the comparability ratio for cause of death i (Ci) is cal- Classification, and Retrieval (MICAR) system makes code assign- culated as: ments to each entity on the death certificate based on the presence or absence of cause-of-death terms and how these are positioned Di,ICD –10 Ci = relative to one another. The Automated Classification of Medical Di,ICD –9 Entities (ACME) program uses the resulting entity codes to select the where Di,ICD–10 is the number of deaths due to cause i classified by underlying cause of death based on a set of decision tables ICD–10 and Di,ICD–9 is the number of deaths due to cause i classified consistent with the ICD selection rules (11–13). See Technical notes by ICD–9. A comparability ratio of 1.00 indicates that the same for more detail regarding the application of the automated coding number of deaths was assigned to cause i under both ICD–9 and systems. ICD–10 denoting no net effect of ICD–10 on that particular cause of Not all of the 1996 mortality records could be classified by ICD–10 death. A ratio showing perfect correspondence between the two using the automated systems. Some records (about 20 percent of the revisions does not necessarily indicate that the cause was totally total 2,314,690) were rejected by the automated systems. This required unaffected by ICD–10, but merely that any increases in the allocation manual intervention to obtain a valid underlying cause of death. More to cause i were completely offset by decreases in the allocation to than 450,000 rejects were initially excluded from the sample and set cause i. aside for manual coding. Of these, a total of 13,703 were manually A comparability ratio less than 1.00 results from fewer deaths coded in time for the release of this report. Records containing mention being classified to cause i under ICD–10 compared with the comparable of certain causes of death are routinely rejected by the automated 6 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 systems, either because the rules governing these causes have not statistics (16). The List of 130 Selected Causes of Infant Death been programmed or because records with mention of these causes (ICD–10 infant cause list) was developed for the analysis of infant require careful examination. For example, these include rare causes, mortality and for ranking leading causes of infant death. It is similar to maternal deaths, some external causes, and deaths due to compli- and replaces the List of 61 Selected Causes of Infant Death (ICD–9 cations of surgeries and adverse effects of drugs. Data for these causes infant cause list) (16). Although the names of the lists imply 113 and of death are sparse in the comparability file. Consequently, the pre- 130 causes of death, these figures refer to the number of mutually liminary data regarding comparability for these causes are not shown. exclusive categories in each list. In reality, the 113-cause list contains Records are also rejected when the software is unable to recognize the a total of 135 cause-of-death categories; the infant list contains 158 terminology listed on the death certificate. total categories. In addition, approximately 40,000 records were excluded from the Comparability ratios are intended to measure the effect of changes sample because the death certificate had initially been submitted with in classification and coding rules. Thus, it is important to avoid poten- cause of death pending investigation or a cause or causes that were tially confounding effects of variation in the tabulation lists used for the later determined to be incorrect. These records were updated in the two revisions. For example, an important cause-of-death category in final mortality file as amendments were received and thus the ICD–9 the 72-cause list is Accidents and adverse effects. However, in the underlying cause is classified correctly. However, updates were not 113-cause list, accidents are categorized separately from the adverse made to the original MICAR input files, and, as a result, ICD–10 effects. A comparability ratio for accidents using the 113-cause list classifications reflect an unknown or incorrect underlying cause. These category in the numerator and the 72-cause list category in the denomi- records are, in almost all cases, external causes (particularly suicides nator reflects not only the changes in classification and coding rules, and homicides) when classified by ICD–9 and are likely to be so but also the decrease in allocation because the deaths from adverse classified in ICD–10. It is possible that some records were overlooked effects are missing in the 113-cause list category. Further, because of that should be excluded from the sample due to a pending certification. the greater detail in the ICD–10 113-cause list, many categories are This should be a very small number overall and should not substantially not included in the 72-cause list (e.g., Salmonella infections, Malignant affect the comparability ratios presented. neoplasm of pancreas, and Alcoholic liver disease). As a result, under Exclusion of the above-mentioned records is unlikely to result in these circumstances, comparability ratios for the causes of death seriously biased estimates of the comparability ratios presented in this added in ICD–10 cannot be calculated. To address these problems, report. Although these records are excluded on a nonrandom basis and ICD–9 codes were assigned to the 113 causes of death in the 113- may not be fully representative of the cause-of-death distribution in the cause list. The same has been done for the infant cause list (i.e., ICD–9 1996 mortality file, the comparability ratios shown in tables 1 and 2 are codes were assigned to the 130 causes of death in the ICD–10 infant indicative of the distribution of deaths by cause of death across the cause list). Table C shows the 113-cause list (135 categories) with Ninth and Tenth Revisions rather than the simple cause-of-death comparable ICD–10 and ICD–9 codes. Table D shows the 130-cause distribution in the mortality file. Generally, for most causes, the cause- list for infant deaths (158 categories) with comparable ICD–10 and of-death distribution of the records across revisions should not be ICD–9 codes. substantially different for the excluded records than for those included in the sample. For example, the probability that a record is classified as heart disease in both ICD–9 and ICD–10 should be similar whether Results the record is included or excluded from the sample. Thus, the sample The following section highlights and explains comparability is assumed to be representative of the distribution of deaths by cause results for major causes of death presented in tables 1 and 2. Table 1 across revisions and is treated as a simple random sample. Even so, shows comparability ratios for the 113-cause list. Table 2 shows for some causes of death, it is possible that the distribution across comparability ratios for the 130-infant-cause list. Ratios are shown revisions may be different for the excluded records. Thus, comparability only for those causes in these lists that are deemed reliable. (See ratios for these causes may be subject to some sampling bias. Where Technical notes for a discussion of the criteria for determining the it is obvious that a cause of death is not adequately represented in the reliability of comparability ratios.) These tables also show the number comparability file, a comparability ratio for that cause is not shown. of deaths in the comparability file classified by ICD–10 and ICD–9, Standard errors for the comparability ratios are calculated using PROC standard errors and 95-percent confidence intervals for comparability RATIO, a procedure in the SUDAAN statistical software designed to ratios. Comparability ratios for major causes of death in the United calculate standard errors for the ratio of two random variables (14). See States are discussed below. Explanations for the magnitude of the Technical notes for more detail about the calculation of the standard comparability ratios are also provided. error of the comparability ratio. Diseases of heart Cause-of-death tabulation lists The comparability ratio for Diseases of heart (113-list number Two of the tabulation lists developed by NCHS for use with 054) is 0.9858, denoting a nearly 1.5 percent net decrease in the mortality data classified by ICD–10 are used to tabulate the compa- allocation of heart disease as the underlying cause of death. Although rability data in this report (15). The List of 113 Selected Causes of it is a small shift in percent, it is a substantial shift in the absolute Death (113-cause list) was developed for the general analysis of number of deaths (8,841). The net decrease is the result primarily of ICD–10 mortality statistics and for ranking leading causes of death. It shifts away from heart diseases to other causes of death due to the is similar in structure and content and replaces the List of 72 change in Rule A. Under Rule A in ICD–10, cardiac arrest is treated Selected Causes of Death (72-cause list) used for ICD–9 mortality as an ill-defined condition. Thus, it is ignored in the selection of the Table C. Comparable category numbers for 113 selected causes of death according to the Ninth and Tenth Revisions, International Classification of Diseases Category codes according to List number Cause of death ICD–101 ICD–91

001 Salmonella infections...... A01–A02 002–003 002 Shigellosis and amebiasis ...... A03,A06 004,006 003 Certain other intestinal infections...... A04,A07–A09 007–009 004 Tuberculosis ...... A16–A19 010–018 005 Respiratory tuberculosis...... A16 010–012 006 Other tuberculosis ...... A17–A19 013–018 007 Whooping cough ...... A37 033 008 Scarlet fever and erysipelas ...... A38,A46 034.1–035 009 Meningococcal infection ...... A39 036 010 Septicemia...... A40–A41 038 011 Syphilis...... A50–A53 090–097 012 Acute poliomyelitis ...... A80 045 013 Arthropod–borne viral encephalitis ...... A83–A84,A85.2 062–064 014 Measles ...... B05 055 015 Viral hepatitis ...... B15–B19 070 016 Human immunodeﬁciency virus (HIV) disease ...... B20–B24 042–044 017 Malaria ...... B50–B54 084 018 Other and unspeciﬁed infectious and parasitic diseases and their sequelae . . . A00,A05,A20–A36,A42–A44,A48–A49,A54– 001,005,020–032,037,039–041,046–054,056– A79,A81–A82,A85.0–A85.1,A85.8, 061,065–066,071–083,085–088,098–134,136, A86–B04,B06–B09,B25–B49,B55–B99 139,771.3 019 Malignant neoplasms...... C00–C97 140–208 020 Malignant neoplasms of lip, oral cavity and pharynx ...... C00–C14 140–149 021 Malignant neoplasm of esophagus ...... C15 150 022 Malignant neoplasm of stomach ...... C16 151 023 Malignant neoplasms of colon, rectum and anus ...... C18–C21 153–154 024 Malignant neoplasms of liver and intrahepatic bile ducts ...... C22 155

025 Malignant neoplasm of pancreas ...... C25 157 2001 18, May 2, No. 49, Vol. Report, Statistics Vital National 026 Malignant neoplasm of larynx ...... C32 161 027 Malignant neoplasms of trachea, bronchus and lung...... C33–C34 162 028 Malignant melanoma of skin ...... C43 172 029 Malignant neoplasm of breast ...... C50 174–175 030 Malignant neoplasm of cervix uteri ...... C53 180 031 Malignant neoplasms of corpus uteri and uterus, part unspeciﬁed...... C54–C55 179,182 032 Malignant neoplasm of ovary ...... C56 183.0 033 Malignant neoplasm of prostate ...... C61 185 034 Malignant neoplasms of kidney and renal pelvis ...... C64–C65 189.0,189.1 035 Malignant neoplasm of bladder ...... C67 188 036 Malignant neoplasms of meninges, brain and other parts of central nervous system ...... C70–C72 191–192 037 Malignant neoplasms of lymphoid, hematopoietic and related tissue ...... C81–C96 200–208 038 Hodgkin’s disease...... C81 201 039 Non-Hodgkin’s lymphoma ...... C82–C85 200,202 040 Leukemia...... C91–C95 204–208 041 Multiple myeloma and immunoproliferative neoplasms ...... C88,C90 203 042 Other and unspeciﬁed malignant neoplasms of lymphoid, hematopoietic and related tissue ...... C96 --- See footnotes at end of table. 7 Table C. Comparable category numbers for 113 selected causes of death according to the Ninth and Tenth Revisions, International Classification of 8 Diseases—Con. 2001 18, May 2, No. 49, Vol. Report, Statistics Vital National Category codes according to List number Cause of death ICD–101 ICD–91

043 All other and unspecified malignant neoplasms ...... C17,C23–C24,C26–C31,C37–C41,C44–C49, 152,156,158–160,163–171,173,181,183.2– C51–C52,C57–C60,C62–C63,C66,C68– 184,186–187,189.2–190,193–199 C69,C73–C80,C97 044 In situ neoplasms, benign neoplasms and neoplasms of uncertain or unknown behavior ...... D00–D48 210–239 045 Anemias ...... D50–D64 280–285 046 Diabetes mellitus ...... E10–E14 250 047 Nutritional deficiencies ...... E40–E64 260–269 048 Malnutrition ...... E40–E46 260–263 049 Other nutritional deficiencies ...... E50–E64 264–269 050 Meningitis ...... G00,G03 320–322 051 Parkinson’s disease ...... G20–G21 332 052 Alzheimer’s disease ...... G30 331.0 053 Major cardiovascular diseases ...... I00–I78 390–434,436–448 054 Diseases of heart ...... I00–I09,I11,I13,I20–I51 390–398,402,404,410–429 055 Acute rheumatic fever and chronic rheumatic heart diseases ...... I00–I09 390–398 056 Hypertensive heart disease ...... I11 402 057 Hypertensive heart and renal disease ...... I13 404 058 Ischemic heart diseases ...... I20–I25 410–414,429.2 059 Acute myocardial infarction ...... I21–I22 410 060 Other acute ischemic heart diseases ...... I24 411 061 Other forms of chronic ischemic heart disease ...... I20,I25 412–414,429.2 062 Atherosclerotic cardiovascular disease, so described ...... I25.0 429.2 063 All other forms of chronic ischemic heart disease ...... I20,I25.1–I25.9 412–414 064 Other heart diseases ...... I26–I51 415–429.1,429.3–429.9 065 Acute and subacute endocarditis ...... I33 421 066 Diseases of pericardium and acute myocarditis ...... I30–I31,I40 420,422–423 067 Heart failure...... I50 428 068 All other forms of heart disease ...... I26–I28,I34–I38,I42–I49,I51 415–417,424–427,429.0–429.1,429.3–429.9 069 Essential (primary) hypertension and hypertensive renal disease ...... I10,I12 401,403 070 Cerebrovascular diseases ...... I60–I69 430–434,436–438 071 Atherosclerosis ...... I70 440 072 Other diseases of circulatory system ...... I71–I78 441–448 073 Aortic aneurysm and dissection ...... I71 441 074 Other diseases of arteries, arterioles and capillaries ...... I72–I78 442–448 075 Other disorders of circulatory system ...... I80–I99 451–459 076 Influenza and pneumonia ...... J10–J18 480–487 077 Influenza ...... J10–J11 487 078 Pneumonia ...... J12–J18 480–486 079 Other acute lower respiratory infections...... J20–J22 466 080 Acute bronchitis and bronchiolitis ...... J20–J21 466 081 Unspecified acute lower respiratory infection ...... J22 --- 082 Chronic lower respiratory diseases...... J40–J47 490–494,496 083 Bronchitis, chronic and unspecified ...... J40–J42 490–491 084 Emphysema ...... J43 492 085 Asthma ...... J45–J46 493 086 Other chronic lower respiratory diseases...... J44,J47 494,496 087 Pneumoconioses and chemical effects ...... J60–J66,J68 500–506 See footnotes at end of table. Table C. Comparable category numbers for 113 selected causes of death according to the Ninth and Tenth Revisions, International Classiﬁcation of Diseases—Con. Category codes according to List number Cause of death ICD–101 ICD–91

088 Pneumonitis due to solids and liquids ...... J69 507 089 Other diseases of respiratory system ...... J00–J06,J30–J39,J67,J70–J98 034.0,460–465,470–478,495,508–519 090 Peptic ulcer ...... K25–K28 531–534 091 Diseases of appendix ...... K35–K38 540–543 092 Hernia ...... K40–K46 550–553 093 Chronic liver disease and cirrhosis ...... K70,K73–K74 571 094 Alcoholic liver disease ...... K70 571.0–571.3 095 Other chronic liver disease and cirrhosis...... K73–K74 571.4–571.9 096 Cholelithiasis and other disorders of gallbladder...... K80–K82 574–575 097 Nephritis, nephrotic syndrome and nephrosis ...... N00–N07,N17–N19,N25–N27 580–589 098 Acute and rapidly progressive nephritic and nephrotic syndrome...... N00–N01,N04 580–581 099 Chronic glomerulonephritis, nephritis and nephropathy not specifiedas.... acute or chronic, and renal sclerosis unspecified...... N02–N03,N05–N07,N26 582–583,587 100 Renal failure ...... N17–N19 584–586 101 Other disorders of kidney...... N25,N27 588–589 102 Infections of kidney...... N10–N12,N13.6,N15.1 590 103 Hyperplasia of prostate ...... N40 600 104 Inflammatory diseases of female pelvic organs ...... N70–N76 614–616 105 Pregnancy, childbirth and the puerperium...... O00–099 630–676 106 Pregnancy with abortive outcome ...... O00–O07 630–639 107 Other complications of pregnancy, childbirth and the puerperium ...... O10–O99 640–676 108 Certain conditions originating in the perinatal period ...... P00–P96 760–771.2,771.4–779 109 Congenital malformations, deformations, and chromosomal abnormalities ..... Q00–Q99 740–759 110 Symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified...... R00–R99 780–799 111 All other diseases (Residual) ...... Residual Residual

112 Accidents (unintentional injuries) ...... V01–X59,Y85–Y86 E800–E869,E880–E929 2001 18, May 2, No. 49, Vol. Report, Statistics Vital National 113 Transport accidents ...... V01–V99,Y85 E800–E848,E929.0,E929.1 114 Motor vehicle accidents ...... V02–V04,V09.0,V09.2,,V12–V14,V19.0– E810–E825 V19.2,V19.4–V19.6,V20–V79,V80.3– V80.5,V81.0–V81.1,V82.0–V82.1,V83–V86,V87.0–V87.8, V88.0–V88.8,V89.0,V89.2 115 Other land transport accidents ...... V01,V05–V06,V09.1,V09.3–V09.9,V10– E800–E807,E826–E829 V11,V15–V18,V19.3,V19.8–V19.9,V80.0– V80.2,V80.6–V80.9,V81.2–V81.9,V82.2– V82.9,V87.9,V88.9,V89.1,V89.3,V89.9 116 Water, air and space, and other and unspecified transport accidents and their sequelae ...... V90–V99,Y85 E830–E848,E929.0,E929.1 117 Nontransport accidents ...... W00–X59,Y86 E850–E869,E880–E928,E929.2–E929.9 118 Falls...... W00–W19 E880–E888 119 Accidental discharge of firearms ...... W32–W34 E922 120 Accidental drowning and submersion ...... W65–W74 E910 121 Accidental exposure to smoke, fire and flames ...... X00–X09 E890–E899 122 Accidental poisoning and exposure to noxious substances...... X40–X49 E850–E869,E924.1 123 Other and unspecified nontransport accidents and their sequelae...... W20–W31,W35–W64,W75–W99,X10– E900–E909,E911–E921,E923–E924.0,E924.8– X39,X50–X59,Y86 E928,E929.2–E929.9 124 Intentional self-harm (suicide)...... X60–X84,Y87.0 E950–E959 125 Intentional self-harm (suicide) by discharge of firearms ...... X72–X74 E955.0–E955.4 See footnotes at end of table. 9 Table C. Comparable category numbers for 113 selected causes of death according to the Ninth and Tenth Revisions, International Classification of 10 Diseases—Con. 2001 18, May 2, No. 49, Vol. Report, Statistics Vital National Category codes according to List number Cause of death ICD–101 ICD–91

126 Intentional self-harm (suicide) by other and unspecified means and their sequelae...... X60–X71,X75–X84,Y87.0 E950–E954,E955.5–E959 127 Assault (homicide) ...... X85–Y09,Y87.1 E960–E969 128 Assault (homicide) by discharge of firearms ...... X93–X95 E965.0–E965.4 129 Assault (homicide) by other and unspecified means and their sequelae .... X85–X92,X96–Y09,Y87.1 E960–E964,E965.5–E969 130 Legal intervention...... Y35,Y89.0 E970–E978 131 Events of undetermined intent ...... Y10–Y34,Y87.2,Y89.9 E980–E989 132 Discharge of firearms, undetermined intent ...... Y22–Y24 E985.0–E985.4 133 Other and unspecified events of undetermined intent and their sequelae.... Y10–Y21,Y25–Y34,Y87.2,Y89.9 E980–E984,E985.5–E989 134 Operations of war and their sequelae ...... Y36,Y89.1 E990–E999 135 Complications of medical and surgical care ...... Y40–Y84,Y88 E870–E879,E930–E949

---Nocomparable category classified by ICD–9 exists. 1ICD–10 is International Classification Diseases, Tenth Revision, and ICD–9isInternational Classification of Diseases, Ninth Revision. Table D. Comparable category numbers for 130 selected causes of infant death according to the Ninth and Tenth Revisions, International Classification of Diseases Category numbers according to List number Cause of death ICD–101 ICD–91

001 Certain infectious and parasitic diseases ...... A00–B99 001–033,034.1–134,136–139,771.3 002 Certain intestinal infectious diseases ...... A00–A08 001–008 003 Diarrhea and gastroenteritis of infectious origin ...... A09 009 004 Tuberculosis ...... A16–A19 010–018 005 Tetanus ...... A33,A35 037,771.3 006 Diptheria ...... A36 032 007 Whooping cough ...... A37 033 008 Meningococcal infection...... A39 036 009 Septicemia ...... A40–A41 038 010 Congenital syphilis...... A50 090 011 Gonococcal infection ...... A54 098 012 Viral diseases ...... A80–B34 042–079 013 Acute poliomyelitis ...... A80 045 014 Varicella (chickenpox) ...... B01 052 015 Measles ...... B05 055 016 Human immunodeficiency virus (HIV) disease ...... B20–B24 042–044 017 Mumps ...... B26 072 018 Other and unspecified viral diseases...... A81–B00,B02–B04,B06–B19,B25,B27–B34 046–051,053–054,056–071,073–079 019 Candidiasis ...... B37 112 020 Malaria ...... B50–B54 084 021 Pneumocystosis ...... B59 136.3 022 All other and unspecified infectious and parasitic diseases ...... A20–A32,A38,A42–A49,A51–A53,A55– 020–031,034.1–035,039–041,080–083,085– A79,B35–B36,B38–B49,B55–B58,B60–B99 088,091–097,099–111,114–134,136.0– 136.2,136.4–139 023 Neoplasms...... C00–D48 140–239 024 Malignant neoplasms ...... C00–C97 140–208 025 Hodgkin’s disease and non-Hodgkin’s lymphomas...... C81–C85 200–202 2001 18, May 2, No. 49, Vol. Report, Statistics Vital National 026 Leukemia...... C91–C95 204–208 027 Other and unspecified malignant neoplasms ...... C00–C80,C88–C90,C96–C97 140–199,203 028 In situ neoplasms, benign neoplasms and neoplasms of uncertain or unknown behavior ...... D00–D48 210–239 029 Diseases of the blood and blood-forming organs and certain disorders involving the immune mechanism...... D50–D89 135,279–289 030 Anemias...... D50–D64 280–285 031 Other diseases of blood and blood-forming organs...... D65–D76 286–289 032 Certain disorders involving the immune mechanism ...... D80–D89 135,279 033 Endocrine, nutritional and metabolic diseases ...... E00–E88 240–278 034 Short stature, not elsewhere classified...... E34.3 259.4 035 Malnutrition and other nutritional deficiencies ...... E40–E64 260–269 036 Cystic fibrosis ...... E84 277.0 037 Volume depletion, disorders of fluid, electrolyte and acid-base balance..... E86–E87 276 038 All other endocrine, nutritional and metabolic diseases ...... E00–E32,E34.0–E34.2,E34.4–E34.9,E65– 240–259.3,259.8–259.9,270–275,277.1–278 E83,E85,E88 039 Diseases of the nervous system ...... G00–G98 320–359,435 040 Meningitis ...... G00,G03 320–322 041 Infantile spinal muscular atrophy, type I (Werdnig-Hoffman)...... G12.0 335.0 042 Infantile cerebral palsy ...... G80 343 043 Anoxic brain damage, not elsewhere classified...... G93.1 348.1 See footnotes at end of table. 11 Table D. Comparable category numbers for 130 selected causes of infant death according to the Ninth and Tenth Revisions, International Classification of 12 Diseases—Con. 2001 18, May 2, No. 49, Vol. Report, Statistics Vital National Category numbers according to List number Cause of death ICD–101 ICD–91

044 Other diseases of nervous system ...... G04,G06–G11,G12.1–G12.9,G20–G72,G81– 323–334,335.1–342,344–348.0,348.2– G92,G93.0,G93.2–G93.9,G95–G98 359,435 045 Diseases of the ear and mastoid process ...... H60–H93 380–389 046 Diseases of the circulatory system...... I00–I99 390–434,436–459 047 Pulmonary heart disease and diseases of pulmonary circulation ...... I26–I28 415–417 048 Pericarditis, endocarditis and myocarditis ...... I30,I33,I40 420–422 049 Cardiomyopathy ...... I42 425 050 Cardiac arrest ...... I46 427.5 051 Cerebrovascular diseases ...... I60–I69 430–434,436–438 052 All other diseases of circulatory system ...... I00–I25,I31,I34–I38,I44–I45,I47–I51,I70–I99 390–414,423–424,426–427.4,427.6–429,440–459 053 Diseases of the respiratory system ...... J00–J98 034.0,460–519 054 Acute upper respiratory infections...... J00–J06 034.0,460–465 055 Influenza and pneumonia...... J10–J18 480–487 056 Influenza ...... J10–J11 487 057 Pneumonia...... J12–J18 480–486 058 Acute bronchitis and acute bronchiolitis ...... J20–J21 466 059 Bronchitis, chronic and unspecified...... J40–J42 490–491 060 Asthma ...... J45–J46 493 061 Pneumonitis due to solids and liquids ...... J69 507 062 Other and unspecified diseases of respiratory system ...... J22,J30–J39,J43–J44,J47–J68,J70–J98 470–479,492,494–506,508–519 063 Diseases of the digestive system ...... K00–K92 520–579 064 Gastritis, duodenitis, and noninfective enteritis and colitis ...... K29,K50–K55 535,555–558 065 Hernia of abdominal cavity and intestinal obstruction without hernia...... K40–K46,K56 550–553,560 066 All other and unspecified diseases of digestive system ...... K00–K28,K30–K38,K57–K92 520–534,536–543,562–579 067 Diseases of the genitourinary system ...... N00–N98 580–629 068 Renal failure and other disorders of kidney ...... N17–N19,N25,N27 584–589 069 Other and unspecified diseases of genitourinary system ...... N00–N15,N20–N23,N26,N28–N98 580–583,590–629 070 Certain conditions originating in the perinatal period ...... P00–P96 760–771.2,771.4–779 071 Newborn affected by maternal factors and by complications of pregnancy, labor and delivery ...... P00–P04 760–763 072 Newborn affected by maternal hypertensive disorders ...... P00.0 760.0 073 Newborn affected by other maternal conditions which may be unrelated to present pregnancy ...... P00.1–P00.9 760.1–760.6,760.8–760.9 074 Newborn affected by maternal complications of pregnancy...... P01 761 075 Newborn affected by incompetent cervix ...... P01.0 761.0 076 Newborn affected by premature rupture of membranes ...... P01.1 761.1 077 Newborn affected by multiple pregnancy ...... P01.5 761.5 078 Newborn affected by other maternal complications of pregnancy...... P01.2–P01.4,P01.6–P01.9 761.2–761.4,761.6–761.9 079 Newborn affected by complications of placenta, cord, and membranes . . . P02 762 080 Newborn affected by complications involving placenta ...... P02.0–P02.3 762.0–762.3 081 Newborn affected by complications involving cord ...... P02.4–P02.6 762.4–762.6 082 Newborn affected by chorioamnionitis...... P02.7 762.7 083 Newborn affected by other and unspecified abnormalities of membranes...... P02.8–P02.9 762.8–762.9 084 Newborn affected by other complications of labor and delivery ...... P03 763.0–763.4,763.6–763.9 085 Newborn affected by noxious influences transmitted via placenta or breast milk ...... P04 760.7, 763.5 086 Disorders related to length of gestation and fetal malnutrition ...... P05–P08 764–766 See footnotes at end of table. Table D. Comparable category numbers for 130 selected causes of infant death according to the Ninth and Tenth Revisions, International Classification of Diseases—Con. Category numbers according to List number Cause of death ICD–101 ICD–91

087 Slow fetal growth and fetal malnutrition ...... P05 764 088 Disorders related to short gestation and low birth weight, not elsewhere classified...... P07 765 089 Extremely low birth weight or extreme immaturity ...... P07.0,P07.2 765.0 090 Other low birth weight or preterm ...... P07.1,P07.3 765.1 091 Disorders related to long gestation and high birth weight ...... P08 766 092 Birth trauma ...... P10–P15 767 093 Intrauterine hypoxia and birth asphyxia ...... P20–P21 768 094 Intrauterine hypoxia ...... P20 768.2–768.4 095 Birth asphyxia ...... P21 768.5–768.9 096 Respiratory distress of newborn ...... P22 769 097 Other respiratory conditions originating in the perinatal period ...... P23–P28 770 098 Congenital pneumonia ...... P23 770.0 099 Neonatal aspiration syndromes ...... P24 770.1 100 Interstitial emphysema and related conditions originating in the perinatal period ...... P25 770.2 101 Pulmonary hemorrhage originating in the perinatal period...... P26 770.3 102 Chronic respiratory disease originating in the perinatal period ...... P27 770.7 103 Atelectasis ...... P28.0–P28.1 770.4–770.5 104 All other respiratory conditions originating in the perinatal period ...... P28.2–P28.9 770.6–770.8 105 Infections specific to the perinatal period ...... P35–P39 771.0–771.2,771.4–771.8 106 Bacterial sepsis of newborn ...... P36 771.8 107 Omphalitis of newborn with or without mild hemorrhage...... P38 771.4 108 All other infections specific to the perinatal period ...... P35,P37,P39 771.0–771.2,771.5–771.7 109 Hemorrhagic and hematological disorders of newborn ...... P50–P61 772–774,776 110 Neonatal hemorrhage ...... P50–P52,P54 772 111 Hemorrhagic disease of newborn ...... P53 776.0 2001 18, May 2, No. 49, Vol. Report, Statistics Vital National 112 Hemolytic disease of newborn due to isoimmunization and other perinatal jaundice...... P55–P59 773–774 113 Hematological disorders ...... P60–P61 776.1–776.9 114 Syndrome of infant of a diabetic mother and neonatal diabetes mellitus ..... P70.0–P70.2 775.0–775.1 115 Necrotizing enterocolitis of newborn ...... P77 777.5 116 Hydrops fetalis not due to hemolytic disease ...... P83.2 778.0 117 Other perinatal conditions ...... P29,P70.3–P76,P78–P81,P83.0–P83.1,P83.3–P96 775.2–775.9,777.0–777.4,777.6–777.9,778.1–779 118 Congenital malformations, deformations and chromosomal abnormalities ..... Q00–Q99 740–759 119 Anencephaly and similar malformations ...... Q00 740 120 Congenital hydrocephalus ...... Q03 742.3 121 Spina bifida...... Q05 741 122 Other congenital malformations of nervous system...... Q01–Q02,Q04,Q06–Q07 742.0–742.2,742.4–742.9 123 Congenital malformations of heart...... Q20–Q24 745–746 124 Other congenital malformations of circulatory system ...... Q25–Q28 747 125 Congenital malformations of respiratory system...... Q30–Q34 748 126 Congenital malformations of digestive system...... Q35–Q45 749–751 127 Congenital malformations of genitourinary system ...... Q50–Q64 752–753 128 Congenital malformations and deformations of musculoskeletal system, limbs and integument ...... Q65–Q85 754–757 129 Down’s syndrome ...... Q90 758.0 130 Edward’s syndrome ...... Q91.0–Q91.3 758.2 See footnotes at end of table. 13 Table D. Comparable category numbers for 130 selected causes of infant death according to the Ninth and Tenth Revisions, International Classification of 14 Diseases—Con. 2001 18, May 2, No. 49, Vol. Report, Statistics Vital National Category numbers according to List number Cause of death ICD–101 ICD–91

131 Patau’s syndrome ...... Q91.4–Q91.7 758.1 132 Other congenital malformations and deformations ...... Q10–Q18,Q86–Q89 743–744,759 133 Other chromosomal abnormalities, not elsewhere classified...... Q92–Q99 758.3–758.9 134 Symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified...... R00–R99 780–799 135 Sudden infant death syndrome ...... R95 798.0 136 Other symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified...... R00–R53,R55–594,R96–R99 780–796,798.1–799 137 All other diseases ...... F01–F99,H00–H57,L00–M99 290–319,360–379,680–739 138 External causes of mortality...... V01–Y89 E800–E999 139 Accidents (unintentional injuries)...... V01–X59 E800–E869,E880–E929 140 Transport accidents...... V01–V99 E800–E848,E920–E929.1 141 Motor vehicle accidents ...... V02–V04,V09.0,V09.2,V12–V14,V19.0–V19.2,V19.4– V19.6,V20–V79,V80.3–V80.5,V81.0–V81.1,V82.0– V82.1,V83–V86,V87.0–V87.8,V88.0– V88.8,V89.0,V89.2 E810–E825 142 Other and unspecified transport accidents ...... V01,V05–V06,V09.1,V09.3–V09.9,V10–V11,V15– V18,V19.3,V19.8,V19.9,V80.0–V80.2,V80.6–V80.9,V81.2– V81.9,V82.2–V82.9,V87.9,V88.9,V89.1,V89.3, V89.9,V90–V99 E800–E807,E826–E848,E929.0–E929.1 143 Falls ...... W00–W19 E880–E888 144 Accidental discharge of firearms...... W32–W34 E922 145 Accidental drowning and submersion ...... W65–W74 E910 146 Accidental suffocation and strangulation in bed ...... W75 E913.0 147 Other accidental suffocation and strangulation ...... W76–W77,W81–W84 E913.1–E913.9 148 Accidental inhalation and ingestion of food or other objects causing obstruction of respiratory tract ...... W78–W80 E911–E912 149 Accidents caused by exposure to smoke, fire and flames...... X00–X09 E890–E899 150 Accidental poisoning and exposure to noxious substances ...... X40–X49 E850–E869,E924.1 151 Other and unspecified accidents ...... W20–W31,W35–W64,W85–W99,X10–X39,X50–X59 E900–E909,E914–E921,E923–E924.0,E924.8–E929 152 Assault (homicide) ...... X85–Y09 E960–E968 153 Assault (homicide) by hanging, strangulation and suffocation...... X91 E963 154 Assault (homicide) by discharge of firearms ...... X93–X95 E965.0–E965.4 155 Neglect, abandonment and other maltreatment syndromes ...... Y06–Y07 E967,E968.4 156 Assault (homicide) by other and unspecified means ...... X85–X90,X92,X96–X99,Y00–Y05,Y08–Y09 E960–E962,E964,E965.5–E966,E968.0– E968.3,968.8–E968.9 157 Complications of medical and surgical care ...... Y40–Y84 E870–E879,E930–E949 158 Other external causes ...... X60–X84,Y10–Y36 E970–E979

1ICD–10 is International Classification of Diseases, Tenth Revision, and ICD–9isInternational Classification of Diseases, Ninth Revision. National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 15 underlying cause of death if another more specific cause is listed on 6 percent in allocation to Cerebrovascular diseases with the imple- the death certificate. Of the heart disease deaths classified by ICD–9, mentation of ICD–10. This increase is due primarily to the change in 98 percent were classified as heart disease in ICD–10. Of the Rule 3. Most of the deaths added to Cerebrovascular diseases in 2 percent not classified as heart disease, most were classified as ICD–10 were classified as pneumonia in ICD–9. Because pneumonia cardiac arrest in ICD–9 (427.5), but were allocated to various other is a direct consequence of Cerebrovascular diseases, Cerebrovas- causes of death in ICD–10. Some deaths were also added to the cular diseases are selected instead of pneumonia as the underlying ICD–10 classification of heart disease from other causes of death. cause of death in ICD–10 when both are listed on the death Many of these are deaths classified in ICD–9 as pneumonia added as certificate. the result of the change in Rule 3. Because pneumonia is a direct consequence of heart disease, heart disease is selected in ICD–10 Chronic lower respiratory diseases as the underlying cause of death when both conditions are listed. The comparability ratio for Chronic lower respiratory diseases Malignant neoplasms (113-list number 082) is 1.0478. The nearly 5-percent increase in allocation to this cause of death is due to the change in Rule 3. The comparability ratio for Malignant neoplasms (113-list number Nearly all of the increase consists of deaths previously coded as 019) is 1.0068. This indicates that the number of deaths due to pneumonia in ICD–9. Chronic lower respiratory diseases are another malignant neoplasms remained stable across revisions. Neverthe- group of causes of death for which pneumonia is a direct conse- less, a substantial number of deaths were classified under Malignant quence. neoplasms in ICD–10 that were not Malignant neoplasms in ICD–9. Among the specified subcategories of Chronic lower respiratory Most of these were classified as pneumonia in ICD–9 and were diseases, comparability ratios show decreases due to the ICD–10 affected by the change in Rule 3. In addition, some deaths shifted out classification. Bronchitis, chronic and unspecified (113-list number of the malignant neoplasm category due to the revision. Most of 083), for example, has a comparability ratio of 0.3935. This indicates these are classified in ICD–10 as HIV or as neoplasms of uncertain a substantial (about 60 percent) decrease when classified by ICD–10. behavior (part of 113-list category In situ neoplasms, benign neo- Although not quite as dramatic, ratios less than 1.0 are also noted for plasms and neoplasms of uncertain or unknown behavior). Emphysema (113-list number 084) and Asthma (113-list number 085). The movement of deaths from malignant neoplasms to HIV results Many deaths coded to bronchitis, emphysema, and asthma in ICD–9 from changes in the interpretation of highly improbable sequences. In are coded in ICD–10 to Other specified chronic obstructive pulmonary ICD–9, most malignant neoplasms listed as due to HIV were considered disease (J44.8), a code that is part of the category Other chronic lower to be improbable sequences. In ICD–10, any malignant neoplasm can respiratory diseases (113-list number 086)—hence the comparability be due to HIV. The shift from malignant neoplasms to neoplasms of ratio of 1.0970 for this category. A code comparable to J44.8 did not uncertain behavior occurs largely as the result of the interpretation of exist in ICD–9. These deaths are affected because of the inclusion on the term ‘‘mass.’’ In ICD–9, presence of the term ‘‘mass’’ was inter- the death certificate of the term ‘‘chronic obstructive pulmonary disease preted to mean a malignancy if reported in specific circumstances with (COPD)—‘‘obstructive’’ being the operative word—with ‘‘bronchitis,’’ other stated malignant neoplasms. The same is not true for ICD–10, ‘‘emphysema,’’ or ‘‘asthma.’’ In ICD–9, if COPD was listed with bron- which treats the term ‘‘mass’’ as a disease of the site of the mass. chitis, the two terms were combined as obstructive chronic bronchitis Some shifts among specified malignant neoplasms occurred with (491.2), which is part of the category Bronchitis, chronic and unspeci- ICD–10 due to the change in the rule governing the selection of the fied. In ICD–10, obstructive chronic bronchitis is coded as J44.8, which primary site. Nearly all the specified malignant neoplasm categories is not part of the category Bronchitis, chronic and unspecified. If COPD show some shifts of deaths into and out of the specified category. For was listed with emphysema or asthma, in ICD–9 emphysema (or example, many deaths due to Malignant neoplasms of trachea, bron- asthma) is preferred as the underlying cause. However, in ICD–10, a chus, and lung (113-list number 027) classified by ICD–9, were clas- linkage has been created so the two terms are combined and coded sified in ICD–10 to more than 15 other specified malignant neoplasm as J44.8. Bronchitis also links with emphysema to become J44.8 in categories, resulting in a comparability ratio of 0.9837. This occurred ICD–10. This combination linkage did not exist in ICD–9. primarily because ICD–10, in contrast with ICD–9, classifies malignant neoplasms of the lung as secondary to many other cancers. All other and unspecified malignant neoplasms (113-list number 043), which has Influenza and pneumonia a comparability ratio of 1.1251, was the category most affected by the A change in the direct sequel rule (Rule 3) is responsible for the change in the rules governing selection of the primary site. The increase comparability ratio of 0.6982 for Influenza and pneumonia (113-list in this category is primarily due to increases in Malignant neoplasms number 076). While Influenza (113-list number 077) is largely of independent (primary) multiple sites (C97), which draws deaths from unaffected by the revision (comparability ratio of 1.0088), Pneumonia a myriad of other specified malignant neoplasms. In ICD–10, selection (113-list number 078) has a comparability ratio of 0.6957. When of the primary site is not determined by order of entry on the death pneumonia is listed on the death certificate with another cause of certificate. Thus, when two primary sites from different organ systems death, and it is obviously a direct consequence of that other cause, are listed, the death is classified to C97. then the other cause is selected as the underlying cause of death. In ICD–10, this rule is applied more broadly than in ICD–9 and specifies Cerebrovascular diseases many more causes for which pneumonia is considered a direct The comparability ratio for Cerebrovascular diseases (113-list consequence. Thus, deaths classified as pneumonia in ICD–9 are number 070) is 1.0588, which indicates a net increase of about classified in ICD–10 to many other causes. The most prominent 16 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 causes are Diseases of heart, Cerebrovascular diseases, Malignant with alcoholism or chronic hepatitis, the death will be classified to neoplasms, Chronic lower respiratory diseases, Septicemia, Malnutri- K70.4. This was not the case in ICD–9. tion, and Chronic liver disease and cirrhosis. Infectious diseases Of all the infectious diseases discussed below, only Septicemia Diabetes mellitus and HIV are currently among leading causes of death. The others— The comparability ratio for Diabetes mellitus (113-list number viral hepatitis, tuberculosis, and syphilis—are very important from a 046) is 1.0082, which indicates a less than 1 percent increase due to public health standpoint and are significantly affected by the imple- ICD–10. This relatively small increase is primarily due to the changes mentation of ICD–10 in Rules A and 3. Some deaths that were coded in ICD–9 to cardiac Septicemia arrest (427.5) are being coded in ICD–10 to diabetes. Because cardiac arrest is treated as an ill-defined condition in ICD–10 under Septicemia (113-list number 010) has a comparability ratio of Rule A, it is ignored in the selection of the underlying cause when 1.1949. A large proportion of this nearly 20 percent increase is due to other better-defined conditions are present. Also, deaths coded in the change in Rule 3 in ICD–10. Under ICD–10, Septicemia is ICD–9 to pneumonia are coded in ICD–10 to diabetes. Complications selected as the underlying cause of death over pneumonia when both of diabetes are among the causes for which pneumonia is an obvious are listed on the death certificate. Also important is a change in the direct consequence. Thus, subsequent to the ICD–10 version of Rule classification of septic shock. Septic shock, coded in ICD–9 as Shock 3, complications of diabetes are selected over pneumonia as the without mention of trauma (785.5) and allocated to the category underlying cause when both conditions are listed on the death Symptoms, signs and abnormal clinical and laboratory findings, not certificate. elsewhere classified, is coded in ICD–10 as Unspecified septicemia (A41.9).

Nephritis, nephrotic syndrome and nephrosis Human immunodeficiency virus (HIV) disease The comparability ratio for HIV (113-list number 016) is 1.0637, Nephritis, nephrotic syndrome and nephrosis (113-list number denoting about 6 percent more deaths due to HIV in ICD–10. The 097) has a comparability ratio of 1.2320. The 23-percent increase in increase in HIV deaths is primarily the result of the change in highly this category is due primarily to changes in the classification of Renal improbable sequences. In ICD–9, only certain specific malignant failure (113-list number 100), which has a comparability ratio of neoplasms could be due to HIV. Malignant neoplasms that could be 1.2949. End-stage renal disease (ESRD), which was classified as an part of a valid sequence with HIV in ICD–9 included Kaposi’s unspecified disorder of the kidney in ICD–9 (593.9) (grouped with All sarcoma (173), Reticulosarcoma (200.0), Burkitt’s tumor (200.2), other diseases), has been reclassified in ICD–10 as End-stage renal immunoblastic sarcoma (200.8), and primary lymphoma of the brain disease (N18.0), a subcategory of Renal failure (N17–N19). This (202.8). If other specified malignant neoplasms were listed as due to results in adding a substantial number of deaths to the Renal failure HIV, the sequence would be considered highly improbable and the and Nephritis, nephrotic syndrome and nephrosis categories. specified malignant neoplasm would be selected as the underlying When applied to years later than 1996, the comparability ratios for cause of death. In ICD–10, any sequence involving a malignant Nephritis, nephrotic syndrome and nephrosis and Renal failure pre- neoplasm due to HIV is considered acceptable, and HIV is selected sented in this report may underestimate the increase in these causes as the underlying cause of death. due to ICD–10. From 1996 through 1999, reporting of ESRD increased The increase in HIV is also partly due to the change in Rule 3 by about 1,900 deaths. This increase disproportionately affects the (pneumonia and the majority of infectious conditions are considered a numerator of the comparability ratio because ESRD is included with direct sequel of HIV). Some records previously classified to viral Renal failure in ICD–10, but not in ICD–9. Thus, the numerator of the hepatitis were added to HIV. Viral hepatitis is a direct sequel of HIV comparability ratio should probably be larger by roughly 1,900 deaths, in ICD–10, but not in ICD–9. When HIV and viral hepatitis are both listed giving a comparability ratio about 1.4 for Renal failure and about 1.3 on the death certificate, HIV is selected as the underlying cause of for Nephritis, nephrotic syndrome and nephrosis. death. When applied to years later than 1996, the comparability ratios for Chronic liver disease and cirrhosis HIV presented in this report may underestimate the increase due to ICD–10. An examination of underlying and multiple cause trends from The comparability ratio for Chronic liver disease and cirrhosis 1996 to 1998 suggest a slightly higher comparability ratio for HIV based (113-list number 093) is 1.0367, denoting a nearly 4-percent increase on 1998 data. due to the implementation of ICD–10. Some of this increase was due to the change in Rule 3. Chronic liver disease is selected as the Viral hepatitis underlying cause of death instead of pneumonia when both are listed Viral hepatitis (113-list number 015) decreased by about 17 per- on the death certificate. The majority of deaths added to Chronic liver cent with the implementation of ICD–10 (comparability ratio of disease and cirrhosis were coded in ICD–9 to Alcohol dependence 0.8343). Most of this decrease is caused by viral hepatitis being syndrome (303) and unspecified chronic hepatitis (573.3). Liver considered as a direct sequel of HIV in ICD–10, but not in ICD–9. failure is linked with alcoholism and chronic hepatitis under the Tenth Thus, when HIV and viral hepatitis are both listed on the death Revision rules to create Alcoholic liver failure (K70.4), a new category certificate, HIV is selected as the underlying cause of death. A part of in ICD–10. Thus, when liver failure is reported on the death certificate the decrease is due to errors in the MICAR system decision tables. National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 17

Deaths are being classified in error to B94.2, Sequelae of viral 1996, resulting in substantial increases in Presenile dementia (290.1) hepatitis. This error will be corrected with the release of the final 1999 from 1996 to 1998. Table E shows the trend in ICD–9-classified Alzhe- mortality data. imer’s disease and Presenile dementia deaths from 1996 to 1998. The number of Alzheimer’s disease deaths increased by about 1,000 deaths Tuberculosis between 1996 and 1997; between 1997 and 1998, the increase was only about 300 deaths. In contrast, the increase in Presenile dementia The comparability ratio for Tuberculosis (113-list number 004) is was more substantial, about 2,000 deaths each year. Thus, the com- 0.8547, a decrease of about 15 percent due to implementation of parability ratio based on 1998 data is probably at least 1.69 (approxi- ICD–10 classification. A small part of this decrease is due to an error mating the ICD–10-classified Alzheimer’s disease deaths by adding the in the MICAR system decision tables. Deaths were classified in error Alzheimer’s disease and Presenile dementia deaths). Assuming pro- to B90.9, Sequelae of tuberculosis. This error will be corrected with portionately similar increases in the ICD–9 classification of Alzheimer’s the release of the final 1999 mortality data. The other part of the disease and Presenile dementia from 1998 to 1999, the comparability decrease is due to a change in the way Lupus, not otherwise ratio based on 1999 data could be as high as 1.8 or 1.9. specified (NOS) is classified. In ICD–9, lupus NOS was assigned a tuberculous code (017.0). In ICD–10, lupus NOS is assigned to Lupus erythematosus (L93). Also, Pneumoconiosis associated with In situ neoplasms, benign neoplasms, and neoplasms of tuberculosis (J65) is a new category in ICD–10 that is included in the uncertain or unknown behavior broader category Pneumoconiosis and chemical effects (113-list For In situ neoplasms, benign neoplasms, and neoplasms of number 087). In ICD–9, deaths with this diagnosis were assigned to uncertain or unknown behavior (113-list number 044), the compara- the tuberculosis category. bility ratio is 1.6744. The increase in this category is due primarily to a classification change involving myelodysplasia. In ICD–9, some of Syphilis these deaths were classified to Diseases of the blood and blood- forming organs (289.8), and others as Other specified anomalies of Syphilis (113-list number 011) has a comparability ratio of spinal cord (742.5). In ICD–10, these deaths are classified to 0.6364, a nearly 37-percent decrease in allocation with the implemen- Myelodysplastic syndrome unspecified (D46.9) or Myeloproliferative tation of ICD–10. Most of the decrease is due to a change in the way syndrome (D47.1). Both are neoplasms of uncertain or unknown late effects (or sequelae) of syphilis are interpreted. In ICD–9, late behavior. Deaths are also being added from various malignant effects or sequelae had to be stated explicitly or the syphilis had to neoplasms due to a change in the interpretation of the term ‘‘mass.’’ be stated as arrested, cured, healed, inactive, old, or quiescent for a In ICD–9, the term ‘‘mass’’ was assumed to be malignant if reported diagnosis involving late effects. In ICD–10, late effects or sequelae of in specific circumstances with other stated malignant neoplasms. In syphilis are determined by duration. If the syphilis or condition or ICD–10, the term ‘‘mass’’ is treated as a disease of the site of the conditions resulting from the syphilis had a duration of 1 year or more mass. then the death is classified to B94.8 which includes sequelae of syphilis. External causes of death

Alzheimer’s disease Accidents (unintentional injuries) The comparability ratio for Alzheimer’s disease (113-list number For Accidents (unintentional injuries) (113-list number 112), the 052) is 1.5536, which indicates a 55 percent increase in Alzheimer’s comparability ratio is 1.0303, which indicates an increase in deaths of disease deaths when classified by ICD–10. In absolute terms, over 3 percent due to the revision. Virtually all of this increase involves 10,000 more deaths were classified to Alzheimer’s disease in ICD–10 shifts from natural causes in ICD–9 to unintentional injuries in than in ICD–9. Nearly all of this increase (about 95 percent) comes ICD–10. Most of these deaths were classified as pneumonia or from deaths classified in ICD–9 as Presenile dementia (290.1). In cardiac arrest in ICD–9, and were changed to unintentional injuries ICD–9,adefinitive diagnosis was required for classification as due to the change in Rule 3 and the change in Rule A, respectively. Alzheimer’s disease. Terms such as ‘‘Alzheimer’s-type dementia’’ or In these cases, the unintentional injury was specified in Part II of the ‘‘Alzheimer’s dementia’’ were classified as Presenile dementia rather death certificate and the pneumonia or cardiac arrest specified in Part than Alzheimer’s disease. In addition, in ICD–9, if an unspecified I. In ICD–9, the conditions reported in Part I are selected as the chronic organic psychotic condition (294.9) is mentioned with Alzhe- underlying cause of death. In ICD–10, pneumonia is a direct imer’s disease, the two conditions form a linkage and are coded to 290.1. Under ICD–10, this linkage does not exist, strictures regarding definitive diagnosis are relaxed, and, thus any mention of Alzheimer’s Table E. Number of deaths due to Alzheimer’s disease is classified as Alzheimer’s disease (G30). This involves the reclas- and Presenile dementia: United States, 1996–98 sification of nearly all cases of Presenile dementia to Alzheimer’s Year disease. The application of the comparability ratio presented for Alzhe- Cause of death (ICD–9 code)1 1996 1997 1998 imer’s disease to years later than 1996 may be problematic (i.e., it may Alzheimer’s disease...... (331.0) 21,397 22,475 22,725 substantially underestimate the increase in Alzheimer’s due to imple- Presenile dementia ...... (290.1) 11,877 13,622 15,672 mentation of the ICD–10 classification). Increases in the reporting of Alzheimer’s-type dementia as a cause of death have occurred since 1ICD–9isInternational Classification of Diseases, Ninth Edition. 18 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 consequence of an unintentional injury. Thus, the unintentional injury the presence of other better-defined conditions. In ICD–10, SIDS is is selected as the underlying cause even if the duration of the injury not considered to be ill-defined. Thus, in ICD–10, SIDS may be is less than one year. In ICD–9, the injury had to occur less than 4 selected as the underlying cause of death even when other condi- weeks before death. Cardiac arrest is treated as an ill-defined tions are listed on the death certificate. condition in ICD–10 and thus, is ignored in the selection of underlying cause, giving preference to the injury listed in Part II. Certain conditions originating in the perinatal period Specific unintentional injuries that merit attention include Motor vehicle accidents (113-list number 114) and Falls (113-list number 118). Six subcategories of Certain conditions originating in the peri- The preliminary comparability ratio for Motor vehicle accidents is natal period are consistently ranked among the leading causes of 0.8527. The reason for this 15-percent decrease is that, in ICD–10, it infant death. They are Disorders related to short gestation and low must be explicit that the injury involved a ‘‘motor’’ vehicle. In ICD–9, birthweight; Newborn affected by maternal complications of preg- in the absence of the term ‘‘motor’’ or when a vehicle accident was nancy; Respiratory distress of newborn; Newborn affected by compli- reported as occurring on a highway or road, the assumption was to cations of placenta, cord, and membranes; Bacterial sepsis of classify the accident as involving a motor vehicle. The ICD–10 con- newborn; and Intrauterine hypoxia and birth asphyxia. The compara- vention does not allow this assumption and classifies such accidents bility ratio for Disorders related to short gestation and low birth weight as involving unspecified vehicles (categorized in ICD–10 as Other land (130-list number 088) is 1.1060. Most of the nearly 11-percent transport accidents). However, for U.S. data, it has been decided that increase is due to the change in Rule A. In ICD–9, these deaths were if the accident occurred on a highway or road, classification to Motor classified to Other respiratory problems after birth (770.8), which vehicle accident is appropriate. This change in classification is effective includes respiratory failure of newborn. In ICD–10, Respiratory failure in the United States with the release of preliminary 1999 mortality data of newborn is treated as an ill-defined condition and is ignored in the and results in a revised comparability ratio for Motor vehicle accidents presence of other better-defined causes. Disorders related to short of 0.9754. This ratio is only applicable to data in which the aforemen- gestation and low birth weight are therefore selected as the under- tioned classification change was implemented; some States may have lying cause of death. released data that does not include this change. Newborn affected by maternal complications of pregnancy (130- Falls decreased about 16 percent with the implementation of list number 074) has a comparability ratio of 1.0295, indicating a ICD–10 (comparability ratio of 0.8409). This decrease is the result of 3-percent increase due to the revision. Deaths added to this category a change in the treatment of unspecified fractures. In ICD–9, if the term were classified in ICD–9 as Other respiratory problems after birth ‘‘fracture’’ is listed on the death certificate without mention of an external (770.8) and Agenesis, hypoplasia, and dysplasia of lung (748.5). cause, the death is classified to Fracture, cause unspecified (E887), Deaths due to Other respiratory problems after birth in ICD–9 are which is grouped with Accidental falls (E880–E888). In ICD–10, a fall classified in ICD–10 to maternal complications due to the change in is not assumed to be responsible for the unspecified fracture, and the Rule A. In addition, in ICD–10, when hypoplasia or dysplasia of lung death is classified to Exposure to unspecified factor (X59), which while is mentioned on the death certificate with prematurity or short gestation, classified as an unintentional injury is not classified as a fall. they form a linkage and the appropriate classification is Primary atelectasis of newborn (P28.0) rather than Hypoplasia and dysplasia Intentional self-harm (suicide) and Assault (homicide) of lung (Q33.6). Maternal complications of pregnancy can be due to Comparability ratios for Intentional self-harm (suicide) (113-list P28.0, but not to Q33.6. Thus, the complication listed is selected as number 124) and Assault (homicide) (113-list number 127) are very the underlying cause of death. close to 1.0. Only a very few deaths were not classified consistently The comparability ratio for Respiratory distress of newborn (130- in these categories. It is not yet clear whether these inconsistencies list number 096) is 1.0257. The 3-percent increase in this category is are real or whether they are records that were unable to be identified the result of the change in Rule A, which shifted deaths from Other as pending amendment. Regardless, the shifts are small enough that respiratory problems after birth (770.8) to Respiratory distress of new- it can be concluded that the revision does not substantially affect born. In ICD–10, respiratory failures are classified as ill-defined con- mortality patterns for suicide or homicide. ditions and ignored in favor of better-defined conditions such as respiratory distress when listed on the death certificate. Infant deaths Newborn affected by complications of placenta, cord, and membranes (130-list number 079) increased by nearly 5 percent (compa- Major causes of death unique to infants are discussed below. rability ratio of 1.0470) due to the implementation of ICD–10. The Comparability ratios for these causes are shown in table 2. Other increase is primarily due to the shifting of deaths from two causes in major infant causes not discussed below, such as unintentional ICD–9, Cerebrovascular diseases, and Respiratory distress of new- injuries and Influenza and pneumonia have comparability patterns born. In ICD–9, infant deaths involving Hypoxic ischemic encephal- similar to those for noninfants. opathy (a cerebrovascular disease coded to 437.1) were not assigned a newborn ICD code. Because the Ninth Revision rules did not allow Sudden infant death syndrome 437.1 to be due to any cause in the newborn chapter, 437.1 was selected as the underlying cause of death. In ICD–10, for infants, The comparability ratio for Sudden infant death syndrome Hypoxic ischemic encephalopathy is classified as Unspecified brain (SIDS) (130-list number 135) is 1.0362. The nearly 4-percent damage due to birth injury (P11.2). Under ICD–10 rules, P11.2 will be increase in SIDS across revisions is due to a change in Rule A. In due to almost any other cause in the newborn chapter including ICD–9, SIDS was treated as an ill-defined condition and ignored in National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 19 complications of placenta, cord, and membranes. In these cases, the requires an enormous investment of national resources to revise latter is selected as the underlying cause of death. Due to a classi- software, training, publications, edit procedures, etc. In the United fication change with ICD–10, Respiratory distress of newborn (P22.0) States, implementation of ICD–10 involved a team of medical can be due to complications of placenta, cord, and membranes. In epidemiologists, statisticians, programmers, nosologists, and systems ICD–9, this sequence was not valid and Respiratory distress of new- analysts over 7 years to complete the task. born would have been selected as the underlying cause of death. Statistically, the new revision and resulting changes in classifi- Bacterial sepsis of newborn (130-list number 106) has a com- cation and rules for selecting the underlying cause of death have parability ratio of 0.9144. The nearly 9-percent drop in this category is important implications for the analysis of mortality trends by cause of primarily the result of system errors in the ICD–10 software (both ACME death. Preliminary comparability ratios by cause of death presented in and MICAR). This resulted in shifts to causes of death other than this report indicate the extent of discontinuities in cause-of-death trends Bacterial sepsis of newborn. This error will be fixed with the 2001 from 1998 to 1999. For some of these causes (e.g., Septicemia, mortality data. When it is corrected, Bacterial sepsis of newborn is likely Influenza and pneumonia, Alzheimer’s disease, and Nephritis, neph- to increase due to the implementation of ICD–10. rotic syndrome, and nephrosis) the discontinuity in trend will be sub- The comparability ratio for Intrauterine hypoxia and birth asphyxia stantial. (130-list number 093) is 1.4477. The 45-percent increase in this cat- New revisions can create substantially changed profiles of health egory is primarily the result of the change in classification involving problems and disease burdens, which are often measured using mor- deaths classified in ICD–9 as Hypoxic ischemic encephalopathy tality data. An example is Healthy People 2010, which uses mortality (437.1). For infants, this cause is reclassified in ICD–10 as Unspecified data to measure health status in a broad range of areas including brain damage due to birth injury (P11.2), which may be due to almost cardiovascular disease, cancer, infant mortality, and diabetes (17). any other cause in the newborn chapter, including birth asphyxia. In Some interesting changes in the overall health profile of the United ICD–9, 437.1 could not be due to causes in the newborn chapter. The States due to the implementation of ICD–10 are shown in table F. This change in Rule A is also important. In ICD–9, categories Other res- is a comparison of the 10 leading causes of death in the United States piratory problems after birth (770.8) and Other and ill-defined conditions for 1996 based on ICD–9 and ICD–10 (10). The top five causes of death originating in the perinatal period (779.8) are not considered to be have the same rank (note that the titles for some causes have changed ill-defined. In ICD–10, these conditions are ill-defined and ignored in between revisions). In contrast, some notable changes in rank occurred favor of Intrauterine hypoxia and birth asphyxia in the selection of the for leading causes 6 through 10. Influenza and pneumonia and Dia- underlying cause of death. betes, ranked 6th and 7th respectively, when classified by ICD–9, switch places when classified by ICD–10. Alzheimer’s disease, pre- Congenital malformations, deformations, and chromosomal viously not ranked in the top 10, is ranked 8th when classified by abnormalities ICD–10. HIV infection and Suicide drop from 8th and 9th to 9th and 10th, respectively. Chronic liver disease and cirrhosis, ranked 10th The comparability ratio for Congenital malformations, deforma- under ICD–9, drop out of the top 10 under ICD–10. tions and chromosomal abnormalities is 0.9064. The nearly 10- percent decrease in deaths from this category is primarily the result Application of comparability ratios of shifts away from Congenital malformations of the respiratory The preliminary comparability ratios presented in this report may system (130-list number 125), particularly Hypoplasia and dysplasia be used as factors for adjusting cause-of-death-specific-mortality of lung, to Primary atelactasis of newborn (P28.0). In ICD–10, when statistics classified by ICD–9 as comparable with cause-specific hypoplasia or dysplasia of lung is mentioned on the death certificate mortality statistics classified by ICD–10. However, this should be with prematurity or short gestation, the appropriate classification is done with caution and an awareness of the limitations of the study. Primary atelectasis of newborn rather than Hypoplasia or dysplasia of The applicability of the preliminary comparability ratios for mea- lung (Q33.6). In ICD–9, no such linkage existed, and the congenital suring discontinuities in cause-of-death trends depends on two impor- anomaly would be the preferred underlying cause. tant factors. First, although sampling bias is generally unlikely as noted Another important classification change involves a shift from Con- in the Methods section of this report, there is potential bias for some genital anomalies of the diaphragm (756.6) in ICD–9 to Diaphragmatic causes of death in the sample selection. Comparability ratios are not hernia without obstruction or gangrene (K44.9) in ICD–10. In ICD–9, shown for causes that are clearly poorly represented in the compa- an infant diagnosis of diaphragmatic hernia is classified by rule as a rability file. Second, the use of 1996 mortality data as the benchmark congenital anomaly even if the congenital anomaly is not explicitly for the comparability file may also introduce bias in comparability ratios stated. In ICD–10, no such rule exists and diaphragmatic hernia is for some causes of death. Such bias is particularly notable for HIV, classified to a K code. Alzheimer’s disease, and Nephritis, nephrotic syndrome, and nephrosis. Because of changes in the composition of the components of Discussion and conclusion these causes since 1996, comparability ratios for these causes, based on 1996 data, are less applicable for measuring the discontinuity in trends from 1998 to 1999. Research to better quantify comparability Impact of new revision ratios for these causes is ongoing. Periodic revision of the ICD is essential to stay abreast of The number of deaths or the death rate for a particular cause of advances in medical science and changes in medical terminology. death classified by ICD–9 is adjusted by multiplying the relevant statistic However, implementation of a new revision is disruptive in both by the comparability ratio for that cause. The following formula can be institutional and statistical terms. Institutionally, revision of the ICD used to modify the number of deaths for cause i classified by ICD–9: 20 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001

Table F. The 10 leading causes of death in 1996 classiﬁed by International Classiﬁcation of Diseases, Ninth and Tenth Revisions Rank ICD–91 tites and codes ICD–101 titles and codes

1 Diseases of heart ...... (390–398,402,404–429) Diseases of heart ...... (I00–I09,I11,I13,I20–I51) 2 Malignant neoplasms, including neoplasms of lymphatic and hematopoietic tissues ...... (140–208) Malignant neoplasms ...... (C00–C97) 3 Cerebrovascular diseases ...... (430–438) Cerebrovascular diseases ...... (I60–I69) 4 Chronic obstructive pulmonary diseases and allied conditions .....(490–496) Chronic lower respiratory diseases ...... (J40–J47) 5 Accidents and adverse effects ...... (E800–E949) Accidents (unintentional injuries) ...... (V01–X59,Y85–Y86) 6 Pneumonia and influenza ...... (480–487) Diabetes mellitus ...... (E10–E14) 7 Diabetes mellitus ...... (250) Influenza and pneumonia ...... (J10–J18) 8 Human immunodeficiency virus (HIV) infection ...... (*042–*044) Alzheimer’s disease...... (G30) 9 Suicide ...... (E950–E959) Human immunodeficiency virus (HIV) disease ...... (B20–B24) 10 Chronic liver disease and cirrhosis ...... (571) Intentional self-harm (suicide) ...... (X60–X84, Y87.0)

1ICD–9isInternational Classiﬁcation Diseases, Ninth Revision, and ICD–10 is International Classiﬁcation of Diseases, Tenth Revision.

CM data. Theoretically, the classification and coding rules for ICD–10 Di,ICD –9 = Di,ICD –9 c Ci should generally be applied uniformly regardless of demographic char- where Di,ICD–9 is the number of deaths for cause i classified by ICD–9, CM acteristics (except in some cases infants will be treated differently) or D i,ICD−9 is the comparability-modified number of deaths for cause i geographic location. Thus, a single comparability ratio for a particular classified by ICD–9, and Ci is the comparability ratio for cause i. The cause of death should theoretically be adequate. However, in practice, death rate for cause i is adjusted similarly by substituting the rate for the demographic and geographic population subgroups will sometimes the number of deaths. Table G shows mortality data for 1998 and be different in composition (i.e., with regard to their cause-of-death preliminary 1999 for Influenza and pneumonia. Ignoring the imple- distribution). This will result in some demographic and geographic mentation of ICD–10, it appears that Influenza and pneumonia deaths variation in cause-specific comparability ratios. The question, yet to be declined dramatically, by 30.7 percent between 1998 and 1999 (from answered, is whether the variation is sufficient to warrant the use of 91,871 to 64,144). Under the same assumption, a comparison of comparability ratios disaggregated by demographic and geographic age-adjusted death rates shows that the risk of dying from Influenza subgroups. and pneumonia dropped by 32.1 percent from 1998 to 1999 (from With the implementation of ICD–10, a set of mortality trends and 34.6 to 23.5 per 100,000 standard population). However, application patterns will emerge that are discontinuous with those produced under of the comparability ratio shows that nearly all of the decline in ICD–9. Trends for many causes of death and the ranking of leading Influenza and pneumonia mortality is due to the introduction of causes of death will be substantially affected. The comparability ratios ICD–10. Modifying the 1998 figures using the comparability ratio for presented in this report, although preliminary, will serve to estimate the Influenza and pneumonia and comparing them with 1999 data shows extent of the discontinuity and will help to make any analysis of mortality that, taking into account the revision of the ICD, the reduction in change between 1998 and 1999 more accurate and complete. The final number of deaths was 0.7 percent, and the risk of dying from comparability study to be prepared during the coming year will provide Influenza and pneumonia declined by only 2.9 percent. Failure to take more robust estimates for the causes described in this paper and new into account the effects of the implementation of ICD–10 can result in information on causes of death not covered in this preliminary study. a greatly distorted view of changes in mortality risk for some causes of death. Caution should be taken when applying the comparability ratios References presented in this report to age-, race-, sex-, or State-specific mortality 1. World Health Organization. International Statistical Classification of Diseases and Related Health Problems,Tenth Revision. Geneva: World Table G. Deaths and death rates for Influenza and Health Organization. 1992. pneumonia: United States, 1998 and preliminary 1999 2. Faust MM, Dolman AB. Comparability ratios based on mortality [1998 data classified using ICD–9 codes 480–487; preliminary 1999 data classified statistics for the fifth and sixth revisions: United States, 1950. Vital using ICD–10 codes J10–J18] Statistics—Special Reports; vol 51, no 3. Washington: Public Health Percent change 1998–99 Service. 1964. 3. Faust MM, Dolman AB. Comparability of mortality statistics for the sixth Due to and seventh revisions: United States, 1958. Vital Statistics—Special Comparability- change in Reports; vol 51, no 4. Washington: Public Health Service. 1965. modified mortality 1998 19981 1999 Overall risk2 4. Klebba AJ. Comparability of mortality statistics for the seventh and eighth revisions of the International Classification of Diseases, United Deaths ...... 91,871 64,144 63,686 –30.7 –0.7 States. National Center for Health Statistics. Vital Health Stat 2(66). Death rate ...... 34.0 23.7 23.4 –31.2 –1.3 1975. Age-adjusted death 5. Klebba AJ. Estimates of selected comparability ratios based on dual rate...... 34.6 24.2 23.5 –32.1 –2.9 coding of 1976 death certificates by the Eighth and Ninth Revisions of 1Calculated by multiplying 1998 data by the comparability ratio for Influenza and pneumonia of the International Classification of Diseases. Monthly Vital Statistics 0.6982. Report; vol 28 no 11, supp. Hyattsville, Maryland: National Center for 2 Comparing comparability-modified 1998 data to preliminary 1999 data. Health Statistics. 1980. National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 21

6. World Health Organization. International Statistical Classification of List of Detailed Tables Diseases and Related Health Problems, Tenth Revision: Volume II, Instruction Manual. Geneva: World Health Organization. 1992. 1. Estimated comparability ratios for 113 selected causes of 7. Tolson GC, Barnes JM, Gay GA, Kowaleski JL. The 1989 revision of death ...... 22 U.S. standard certificates and reports. National Center for Health 2. Estimated comparability ratios for 130 selected causes of infant Statistics. Vital Health Stat 4(28). 1991. death ...... 25 8. National Center for Health Statistics. Instruction Manual Part 2a: Instructions for Classifying the Underlying Cause of Death. Research Triangle Park: National Center for Health Statistics. 1999. 9. World Health Organization. Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death, Ninth Revision: Volume I. Geneva: World Health Organization. 1977. 10. Peters KD, Kochanek KD, Murphy SL. Deaths: Final Data for 1996. National Vital Statistics Reports; vol 47, no 9. Hyattsville, Maryland: National Center for Health Statistics. 1998. 11. Hetzel AM. U.S. Vital Statistics System: Major activities and develop- ments, 1950–95. Research Triangle Park: National Center for Health Statistics. 1997. 12. National Center for Health Statistics. Instruction Manual Part 2g: Data Entry Instructions for the Mortality Medical Indexing, Classification, and Retrieval System (MICAR). Research Triangle Park: National Center for Health Statistics. 1999. 13. National Center for Health Statistics. Instruction Manual Part 2c: ICD-10 ACME decision tables for classifying underlying causes of death. Hyattsville, Maryland: National Center for Health Statistics. 1999. 14. Shah BV, Barnwell BG, Bieler GS. SUDAAN User’s Manual. Research Triangle Park, NC: Research Triangle Institute. 1996. 15. National Center for Health Statistics. Instruction Manual Part 9: ICD-10 cause of death lists for tabulating mortality statistics, effective 1999. Hyattsville, Maryland: National Center for Health Statistics. 1999. 16. National Center for Health Statistics. Instruction Manual Part 9: ICD-9 cause of death lists for tabulating mortality statistics, effective 1979. Hyattsville, Maryland: National Center for Health Statistics. 1979. 17. U.S. Department of Health and Human Services. Healthy people 2010. 2nd edition. Two volumes. Washington, DC. U.S. Government Printing Office, November 2000. 18. National Center for Health Statistics. Proceedings of the International Collaborative Effort on Automating Mortality Statistics, Volume I. Peters KD, ed. Hyattsville, Maryland. 1999. 19. Shah BV, Folsom RE, LaVange LM, Wheeless SC, Boyle KE, Williams RL. Statistical Methods and Mathematical Algorithms Used in SUDAAN. Research Triangle Park, NC: Research Triangle Institute. 1993. 20. Murphy SL. Deaths: Final data for 1998. National Vital Statistics Reports; vol 48, no 11. Hyattsville, Maryland: National Center for Health Statistics. 2000. 21. Kochanek KD, Smith BL, Anderson RN. Deaths: Preliminary data for 1999. National Vital Statistics Reports; vol 49, no 3. Hyattsville, Maryland: National Center for Health Statistics. 2001. 22. Miniño AM. A guide to State implementation of ICD-10 for mortality, Part II: Applying comparability ratios. Hyattsville, Maryland: National Center for Health Statistics. 2000. http://www.cdc.gov/ nchs/about/major/dvs/icd10des.htm 22 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 Table 1. Estimated comparability ratios for 113 selected causes of death Number of deaths allocated 95 percent according to Estimated Relative confidence limits List comparability Standard standard number Cause of death1 ICD–102 ICD–92 ratio error error Lower Upper

001 Salmonella infections ...... 30 37 0.8108 0.0644 7.9 0.6846 0.9370 002 Shigellosis and amebiasis ...... * * * * * * * 003 Certain other intestinal infections ...... * * * * * * * 004 Tuberculosis ...... 653 764 0.8547 0.0172 2 0.8209 0.8885 005 Respiratory tuberculosis...... 518 572 0.9056 0.0201 2.2 0.8662 0.9450 006 Other tuberculosis ...... 135 192 0.7031 0.0407 5.8 0.6233 0.7830 007 Whooping cough ...... * * * * * * * 008 Scarlet fever and erysipelas...... * * * * * * * 009 Meningococcal infection ...... 221 222 0.9955 0.0149 1.5 0.9663 1.0247 010 Septicemia ...... 21,258 17,791 1.1949 0.0042 0.3 1.1867 1.2030 011 Syphilis ...... 21 33 0.6364 0.1184 18.6 0.4043 0.8685 012 Acute poliomyelitis ...... * * * * * * * 013 Arthropod-borne viral encephalitis ...... * * * * * * * 014 Measles ...... * * * * * * * 015 Viral hepatitis ...... 1,123 1,346 0.8343 0.0120 1.4 0.8109 0.8578 016 Human immunodeficiency virus (HIV) disease ...... 25,089 23,586 1.0637 0.0018 0.2 1.0601 1.0673 017 Malaria ...... * * * * * * * 018 Other and unspecified infectious and parasitic diseases and their sequelae ...... 2,865 2,607 1.0990 0.0154 1.4 1.0688 1.1291 019 Malignant neoplasms ...... 464,688 461,544 1.0068 0.0002 0.0 1.0064 1.0072 020 Malignant neoplasms of lip, oral cavity and pharynx ...... 5,927 6,172 0.9603 0.0040 0.4 0.9525 0.9681 021 Malignant neoplasm of esophagus ...... 9,596 9,630 0.9965 0.0020 0.2 0.9926 1.0003 022 Malignant neoplasm of stomach ...... 11,480 11,408 1.0063 0.0019 0.2 1.0025 1.0101 023 Malignant neoplasms of colon, rectum and anus ...... 48,583 48,619 0.9993 0.0009 0.1 0.9975 1.0010 024 Malignant neoplasms of liver and intrahepatic bile ducts...... 9,732 10,102 0.9634 0.0023 0.2 0.9588 0.9679 025 Malignant neoplasm of pancreas ...... 24,313 24,361 0.9980 0.0009 0.1 0.9963 0.9997 026 Malignant neoplasm of larynx ...... 3,209 3,194 1.0047 0.0053 0.5 0.9943 1.0150 027 Malignant neoplasms of trachea, bronchus and lung ...... 131,750 133,936 0.9837 0.0005 0.1 0.9827 0.9846 028 Malignant melanoma of skin ...... 5,941 6,139 0.9677 0.0032 0.3 0.9614 0.9741 029 Malignant neoplasm of breast ...... 38,102 37,891 1.0056 0.0010 0.1 1.0036 1.0075 030 Malignant neoplasm of cervix uteri ...... 3,753 3,802 0.9871 0.0034 0.3 0.9805 0.9938 031 Malignant neoplasms of corpus uteri and uterus, part unspecified . . . 5,318 5,183 1.0260 0.0040 0.4 1.0182 1.0339 032 Malignant neoplasm of ovary...... 11,292 11,344 0.9954 0.0016 0.2 0.9923 0.9985 033 Malignant neoplasm of prostate ...... 30,672 30,267 1.0134 0.0015 0.1 1.0105 1.0162 034 Malignant neoplasms of kidney and renal pelvis ...... 9,521 9,521 1.0000 0.0022 0.2 0.9957 1.0043 035 Malignant neoplasm of bladder ...... 9,563 9,594 0.9968 0.0026 0.3 0.9916 1.0019 036 Malignant neoplasms of meninges, brain and other parts of central nervous system ...... 10,039 10,359 0.9691 0.0025 0.3 0.9642 0.9740 037 Malignant neoplasms of lymphoid, hematopoietic and related tissue ...... 44,715 44,530 1.0042 0.0012 0.1 1.0019 1.0064 038 Hodgkin’s disease ...... 1,021 1,036 0.9855 0.0089 0.9 0.9680 1.0030 039 Non-Hodgkin’s lymphoma ...... 17,924 18,326 0.9781 0.0018 0.2 0.9745 0.9817 040 Leukemia ...... 16,600 16,405 1.0119 0.0019 0.2 1.0083 1.0155 041 Multiple myeloma and immunoproliferative neoplasms...... 9,099 8,763 1.0383 0.0030 0.3 1.0324 1.0443 042 Other and unspecified malignant neoplasms of lymphoid, hematopoietic and related tissue ...... * * * * * * * 043 All other and unspecified malignant neoplasms ...... 51,182 45,492 1.1251 0.0021 0.2 1.1210 1.1292 044 In situ neoplasms, benign neoplasms and neoplasms of uncertain or unknown behavior ...... 9,263 5,532 1.6744 0.0164 1.0 1.6422 1.7067 045 Anemias ...... 3,059 3,200 0.9559 0.0077 0.8 0.9409 0.9710 046 Diabetes mellitus ...... 48,636 48,242 1.0082 0.0011 0.1 1.0060 1.0103 047 Nutritional deficiencies ...... 3,215 2,763 1.1636 0.0165 1.4 1.1312 1.1960 048 Malnutrition ...... 2,607 2,665 0.9782 0.0151 1.5 0.9487 1.0078 049 Other nutritional deficiencies ...... 608 98 6.2041 0.5961 9.6 5.0358 7.3724 050 Meningitis ...... 592 584 1.0137 0.0136 1.3 0.9871 1.0403 051 Parkinson’s disease...... 10,404 10,392 1.0012 0.0028 0.3 0.9956 1.0067 052 Alzheimer’s disease...... 29,707 19,121 1.5536 0.0071 0.5 1.5398 1.5675 053 Major cardiovascular diseases ...... 796,919 798,435 0.9981 0.0002 0.0 0.9977 0.9985 054 Diseases of heart ...... 615,564 624,405 0.9858 0.0002 0.0 0.9854 0.9863 055 Acute rheumatic fever and chronic rheumatic heart diseases .... 2,446 2,980 0.8208 0.0089 1.1 0.8034 0.8382 056 Hypertensive heart disease ...... 17,322 21,577 0.8028 0.0028 0.3 0.7973 0.8083 057 Hypertensive heart and renal disease ...... 2,170 2,027 1.0705 0.0160 1.5 1.0392 1.1019 058 Ischemic heart diseases ...... 466,459 466,935 0.9990 0.0002 0.0 0.9985 0.9994 059 Acute myocardial infarction ...... 178,125 180,169 0.9887 0.0003 0.0 0.9880 0.9893 060 Other acute ischemic heart diseases ...... 2,667 2,638 1.0110 0.0117 1.2 0.9880 1.0340 061 Other forms of chronic ischemic heart disease ...... 285,667 284,128 1.0054 0.0004 0.0 1.0046 1.0062 See footnotes at end of table. National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 23 Table 1. Estimated comparability ratios for 113 selected causes of death—Con. Number of deaths allocated 95 percent according to Estimated Relative confidence limits List comparability Standard standard number Cause of death1 ICD–102 ICD–92 ratio error error Lower Upper

062 Atherosclerotic cardiovascular disease, so described...... 64,354 61,362 1.0488 0.0016 0.2 1.0456 1.0519 063 All other forms of chronic ischemic heart disease ...... 221,313 222,766 0.9935 0.0004 0.0 0.9927 0.9942 064 Other heart diseases ...... 127,167 130,886 0.9716 0.0010 0.1 0.9696 0.9736 065 Acute and subacute endocarditis...... 552 554 0.9964 0.0137 1.4 0.9695 1.0233 066 Diseases of pericardium and acute myocarditis ...... 489 475 1.0295 0.0160 1.6 0.9981 1.0608 067 Heart failure ...... 44,297 42,554 1.0410 0.0013 0.1 1.0384 1.0435 068 All other forms of heart disease ...... 81,829 87,303 0.9373 0.0014 0.2 0.9345 0.9401 069 Essential (primary) hypertension and hypertensive renal disease.... 11,958 10,684 1.1192 0.0050 0.4 1.1094 1.1291 070 Cerebrovascular diseases...... 137,264 129,640 1.0588 0.0008 0.1 1.0572 1.0604 071 Atherosclerosis ...... 13,894 14,417 0.9637 0.0025 0.3 0.9588 0.9686 072 Other diseases of circulatory system ...... 18,239 19,289 0.9456 0.0021 0.2 0.9414 0.9498 073 Aortic aneurysm and dissection ...... 12,216 12,201 1.0012 0.0010 0.1 0.9992 1.0032 074 Other diseases of arteries, arterioles and capillaries...... 6,023 7,088 0.8497 0.0053 0.6 0.8394 0.8601 075 Other disorders of circulatory system ...... 2,984 2,899 1.0293 0.0172 1.7 0.9956 1.0631 076 Influenza and pneumonia ...... 50,526 72,371 0.6982 0.0018 0.3 0.6947 0.7016 077 Influenza...... 572 567 1.0088 0.0073 0.7 0.9945 1.0231 078 Pneumonia ...... 49,954 71,804 0.6957 0.0018 0.3 0.6922 0.6992 079 Other acute lower respiratory infections ...... 346 355 0.9746 0.0392 4.0 0.8978 1.0515 080 Acute bronchitis and bronchiolitis ...... 265 355 0.7465 0.0264 3.5 0.6947 0.7983 081 Unspecified acute lower respiratory infection...... * * * * * * * 082 Chronic lower respiratory diseases ...... 94,326 90,022 1.0478 0.0009 0.1 1.0460 1.0496 083 Bronchitis, chronic and unspecified...... 913 2,320 0.3935 0.0107 2.7 0.3726 0.4145 084 Emphysema...... 14,369 14,774 0.9726 0.0031 0.3 0.9666 0.9786 085 Asthma ...... 4,217 4,718 0.8938 0.0061 0.7 0.8819 0.9057 086 Other chronic lower respiratory diseases ...... 74,827 68,210 1.0970 0.0014 0.1 1.0943 1.0998 087 Pneumoconioses and chemical effects ...... 860 845 1.0178 0.0099 1.0 0.9983 1.0372 088 Pneumonitis due to solids and liquids ...... 10,183 9,104 1.1185 0.0048 0.4 1.1092 1.1279 089 Other diseases of respiratory system ...... 16,656 14,269 1.1673 0.0052 0.4 1.1572 1.1774 090 Peptic ulcer ...... 3,574 3,686 0.9696 0.0045 0.5 0.9608 0.9784 091 Diseases of appendix ...... 209 202 1.0347 0.0242 2.3 0.9873 1.0820 092 Hernia ...... 658 633 1.0395 0.0154 1.5 1.0094 1.0696 093 Chronic liver disease and cirrhosis...... 21,688 20,920 1.0367 0.0027 0.3 1.0314 1.0420 094 Alcoholic liver disease...... 10,147 9,965 1.0183 0.0050 0.5 1.0085 1.0281 095 Other chronic liver disease and cirrhosis ...... 11,541 10,955 1.0535 0.0041 0.4 1.0454 1.0615 096 Cholelithiasis and other disorders of gallbladder ...... 1,725 1,803 0.9567 0.0060 0.6 0.9450 0.9685 097 Nephritis, nephrotic syndrome and nephrosis...... 24,939 20,242 1.2320 0.0044 0.4 1.2234 1.2407 098 Acute and rapidly progressive nephritic and nephrotic syndrome .... 161 249 0.6466 0.0342 5.3 0.5796 0.7136 099 Chronic glomerulonephritis, nephritis and nephropathy not specified as acute or chronic, and renal sclerosis unspecified...... 468 1,213 0.3858 0.0144 3.7 0.3575 0.4141 100 Renal failure ...... 24,290 18,758 1.2949 0.0050 0.4 1.2852 1.3047 101 Other disorders of kidney ...... 20 22 0.9091 0.0867 9.5 0.7392 1.0790 102 Infections of kidney ...... 731 726 1.0069 0.0144 1.4 0.9786 1.0352 103 Hyperplasia of prostate...... 326 327 0.9969 0.0159 1.6 0.9658 1.0280 104 Inflammatory diseases of female pelvic organs...... 63 64 0.9844 0.0410 4.2 0.9040 1.0648 105 Pregnancy, childbirth and the puerperium ...... * * * * * * * 106 Pregnancy with abortive outcome ...... * * * * * * * 107 Other complications of pregnancy, childbirth and the puerperium.... * * * * * * * 108 Certain conditions originating in the perinatal period ...... 10,184 9,555 1.0658 0.0033 0.3 1.0593 1.0724 109 Congenital malformations, deformations and chromosomal abnormalities ...... 5,950 7,025 0.8470 0.0055 0.6 0.8362 0.8577 110 Symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified...... 16,940 17,732 0.9553 0.0034 0.4 0.9487 0.9620 111 All other diseases (Residual) ...... 109,853 122,107 0.8996 0.0015 0.2 0.8968 0.9025 112 Accidents (unintentional injuries) ...... 31,084 30,163 1.0305 0.0014 0.1 1.0278 1.0333 113 Transport accidents ...... 17,547 17,586 0.9978 0.0006 0.1 0.9966 0.9990 114 Motor vehicle accidents ...... 14,539 17,051 0.8527 0.0027 0.3 0.8473 0.8581 115 Other land transport accidents ...... * * * * * * * 116 Water, air and space, and other and unspecified transport accidents and their sequelae...... 351 347 1.0115 0.0209 2.1 0.9706 1.0525 117 Nontransport accidents ...... 13,537 12,577 1.0763 0.0035 0.3 1.0696 1.0831 118 Falls ...... 5,173 6,152 0.8409 0.0049 0.6 0.8313 0.8505 119 Accidental discharge of firearms ...... 493 466 1.0579 0.0127 1.2 1.0331 1.0828 120 Accidental drowning and submersion...... 283 284 0.9965 0.0127 1.3 0.9716 1.0213 121 Accidental exposure to smoke, fire and flames...... 493 506 0.9743 0.0089 0.9 0.9568 0.9918 122 Accidental poisoning and exposure to noxious substances ...... * * * * * * * 123 Other and unspecified nontransport accidents and their sequelae . . 6,698 4,721 1.4188 0.0123 0.9 1.3947 1.4428 See footnotes at end of table. 24 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 Table 1. Estimated comparability ratios for 113 selected causes of death—Con. Number of deaths allocated 95 percent according to Estimated Relative confidence limits List comparability Standard standard number Cause of death1 ICD–102 ICD–92 ratio error error Lower Upper

124 Intentional self-harm (suicide) ...... 18,352 18,422 0.9962 0.0005 0.0 0.9952 0.9972 125 Intentional self-harm (suicide) by discharge of firearms ...... 14,157 14,183 0.9982 0.0007 0.1 0.9968 0.9996 126 Intentional self-harm (suicide) by other and unspecified means and their sequelae ...... 4,195 4,239 0.9896 0.0023 0.2 0.9850 0.9942 127 Assault (homicide) ...... 12,287 12,308 0.9983 0.0006 0.1 0.9972 0.9994 128 Assault (homicide) by discharge of firearms ...... 8,718 8,745 0.9969 0.0008 0.1 0.9953 0.9985 Assault (homicide) by other and unspecified means and their 129 sequelae...... 3,569 3,563 1.0017 0.0024 0.2 0.9969 1.0064 130 Legal intervention ...... * * * * * * * 131 Events of undetermined intent ...... * * * * * * * 132 Discharge of firearms, undetermined intent ...... * * * * * * * 133 Other and unspecified events of undetermined intent and their sequelae ...... * * * * * * * 134 Operations of war and their sequelae ...... * * * * * * * 135 Complications of medical and surgical care...... * * * * * * *

* Figure does not meet standards of reliability or precision; see Technical notes. 0.0 Quantity more than zero but less than 0.05. 1Based on the Ninth and Tenth Revision categories shown in table B. 2ICD–10 is International Classification of Diseases, Tenth Revision, and ICD-9 is International Classification of Diseases, Ninth Revision. National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 25 Table 2. Estimated comparability ratios for 130 selected causes of infant death Number of deaths allocated 95 percent according to Estimated Relative confidence limits List comparability Standard standard number Cause of death1 ICD–102 ICD–92 ratio error error Lower Upper

001 Certain infectious and parasitic diseases ...... 284 387 0.7339 0.0339 4.6 0.6673 0.8004 002 Certain intestinal infectious diseases ...... * * * * * * * 003 Diarrhea and gastroenteritis of infectious origin ...... – 144 ––––– 004 Tuberculosis...... * * * * * * * 005 Tetanus ...... * * * * * * * 006 Diptheria...... * * * * * * * 007 Whooping cough ...... * * * * * * * 008 Meningococcal infection ...... 25 26 0.9615 0.0377 3.9 0.8876 1.0355 009 Septicemia ...... 167 121 1.3802 0.0713 5.2 1.2403 1.5200 010 Congenital syphilis ...... * * * * * * * 011 Gonococcal infection...... * * * * * * * 012 Viral diseases...... 62 62 1.0000 0.0757 7.6 0.8517 1.1483 013 Acute poliomyelitis...... * * * * * * * 014 Varicella (chickenpox)...... * * * * * * * 015 Measles...... * * * * * * * 016 Human immunodeficiency virus (HIV) disease ...... 23 22 1.0455 0.0465 4.4 0.9544 1.1365 017 Mumps ...... * * * * * * * 018 Other and unspecified viral diseases ...... 35 36 0.9722 0.1255 12.9 0.7262 1.2182 019 Candidiasis ...... * * * * * * * 020 Malaria...... * * * * * * * 021 Pneumocystosis ...... * * * * * * * 022 All other and unspecified infectious and parasitic diseases ...... * * * * * * * 023 Neoplasms ...... 73 72 1.0139 0.0420 4.1 0.9317 1.0961 024 Malignant neoplasms ...... 48 46 1.0435 0.0544 5.2 0.9369 1.1501 025 Hodgkin’s disease and non-Hodgkin’s lymphomas ...... * * * * * * * 026 Leukemia ...... * * * * * * * 027 Other and unspecified malignant neoplasms ...... 30 28 1.0714 0.0906 8.5 0.8939 1.2489 028 In situ neoplasms, benign neoplasms and neoplasms of uncertain or unknown behavior ...... 25 26 0.9615 0.1131 11.8 0.7398 1.1833 029 Diseases of the blood and blood-forming organs and certain disorders involving the immune mechanism ...... 35 50 0.7000 0.0803 11.5 0.5427 0.8573 030 Anemias ...... * * * * * * * 031 Other diseases of blood and blood-forming organs ...... * * * * * * * 032 Certain disorders involving the immune mechanism ...... * * * * * * * 033 Endocrine, nutritional and metabolic diseases ...... 112 129 0.8682 0.0555 6.4 0.7595 0.9770 034 Short stature, not elsewhere classified...... * * * * * * * 035 Malnutrition and other nutritional deficiencies ...... * * * * * * * 036 Cystic fibrosis...... * * * * * * * 037 Volume depletion, disorders of fluid, electrolyte and acid-base balance...... 40 53 0.7547 0.0852 11.3 0.5878 0.9217 038 All other endocrine, nutritional and metabolic diseases...... 64 55 1.1636 0.0809 6.9 1.0051 1.3221 039 Diseases of the nervous system ...... 305 286 1.0664 0.0263 2.5 1.0149 1.1180 040 Meningitis ...... 70 70 1.0000 0.0404 4.0 0.9208 1.0792 041 Infantile spinal muscular atrophy, type I (Werdnig-Hoffman) ...... 47 47 1.0000 0.0521 5.2 0.8978 1.1022 042 Infantile cerebral palsy ...... * * * * * * * 043 Anoxic brain damage, not elsewhere classified...... 29 30 0.9667 0.1269 13.1 0.7179 1.2155 044 Other diseases of nervous system ...... 145 126 1.1508 0.0532 4.6 1.0466 1.2550 045 Diseases of the ear and mastoid process...... * * * * * * * 046 Diseases of the circulatory system ...... 419 587 0.7138 0.0244 3.4 0.6659 0.7617 047 Pulmonary heart disease and diseases of pulmonary circulation .... 138 123 1.1220 0.0447 4.0 1.0342 1.2097 048 Pericarditis, endocarditis and myocarditis...... * * * * * * * 049 Cardiomyopathy ...... 82 84 0.9762 0.0166 1.7 0.9436 1.0088 050 Cardiac arrest...... 25 87 0.2874 0.0508 17.7 0.1878 0.3869 051 Cerebrovascular diseases...... 77 163 0.4724 0.0510 10.8 0.3725 0.5723 052 All other diseases of circulatory system ...... 88 123 0.7154 0.0519 7.3 0.6137 0.8172 053 Diseases of the respiratory system...... 420 516 0.8140 0.0220 2.7 0.7709 0.8570 054 Acute upper respiratory infections ...... * * * * * * * 055 Influenza and pneumonia ...... 231 303 0.7624 0.0261 3.4 0.7112 0.8135 056 Influenza ...... * * * * * * * 057 Pneumonia ...... 224 295 0.7593 0.0266 3.5 0.7072 0.8114 058 Acute bronchitis and acute bronchiolitis...... 33 41 0.8049 0.0758 9.4 0.6563 0.9534 059 Bronchitis, chronic and unspecified...... * * * * * * * 060 Asthma ...... * * * * * * * 061 Pneumonitis due to solids and liquids...... * * * * * * * 062 Other and unspecified diseases of respiratory system ...... 117 127 0.9213 0.0632 6.9 0.7973 1.0452 063 Diseases of the digestive system...... 278 167 1.6647 0.1084 6.5 1.4521 1.8772 See footnotes at end of table. 26 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 Table 2. Estimated comparability ratios for 130 selected causes of infant death—Con. Number of deaths allocated 95 percent according to Estimated Relative confidence limits List comparability Standard standard number Cause of death1 ICD–102 ICD–92 ratio error error Lower Upper

064 Gastritis, duodenitis, and noninfective enteritis and colitis ...... 137 47 2.9149 0.3879 13.3 2.1547 3.6751 065 Hernia of abdominal cavity and intestinal obstruction without hernia . . * * * * * * * 066 All other and unspecified diseases of digestive system ...... 84 86 0.9767 0.0708 7.3 0.8379 1.1156 067 Diseases of the genitourinary system ...... 117 117 1.0000 0.0567 5.7 0.8889 1.1111 068 Renal failure and other disorders of kidney ...... 102 98 1.0408 0.0658 6.3 0.9118 1.1699 069 Other and unspecified diseases of genitourinary system ...... * * * * * * * 070 Certain conditions originating in the perinatal period ...... 10,047 9,495 1.0581 0.0032 0.3 1.0519 1.0643 071 Newborn affected by maternal factors and by complications of pregnancy, labor and delivery...... 1,305 1,256 1.0390 0.0099 1.0 1.0196 1.0585 072 Newborn affected by maternal hypertensive disorders...... 23 22 1.0455 0.0465 4.4 0.9544 1.1365 073 Newborn affected by other maternal conditions which may be unrelated to present pregnancy ...... * * * * * * * 074 Newborn affected by maternal complications of pregnancy ...... 662 643 1.0295 0.0138 1.3 1.0024 1.0567 075 Newborn affected by incompetent cervix ...... 205 201 1.0199 0.0188 1.8 0.9831 1.0567 076 Newborn affected by premature rupture of membranes...... 314 307 1.0228 0.0136 1.3 0.9962 1.0494 077 Newborn affected by multiple pregnancy ...... 104 103 1.0097 0.0507 5.0 0.9103 1.1091 078 Newborn affected by other maternal complications of pregnancy ...... 39 32 1.2188 0.1655 13.6 0.8945 1.5430 079 Newborn affected by complications of placenta, cord, and membranes ...... 579 553 1.0470 0.0128 1.2 1.0219 1.0721 080 Newborn affected by complications involving placenta ...... 306 285 1.0737 0.0174 1.6 1.0395 1.1079 081 Newborn affected by complications involving cord...... * * * * * * * 082 Newborn affected by chorioamnionitis ...... 258 255 1.0118 0.0163 1.6 0.9799 1.0436 083 Newborn affected by other and unspecified abnormalities of membranes ...... * * * * * * * 084 Newborn affected by other complications of labor and delivery . . . 37 20 1.8500 0.3262 17.6 1.2107 2.4893 085 Newborn affected by noxious influences transmitted via placenta or breast milk ...... * * * * * * * 086 Disorders related to length of gestation and fetal malnutrition...... 3,843 3,474 1.1062 0.0064 0.6 1.0936 1.1188 087 Slow fetal growth and fetal malnutrition ...... 34 30 1.1333 0.1004 8.9 0.9366 1.3301 088 Disorders related to short gestation and low birth weight, not elsewhere classified...... 3,809 3,444 1.1060 0.0064 0.6 1.0934 1.1186 089 Extremely low birth weight or extreme immaturity ...... 2,835 2,558 1.1083 0.0079 0.7 1.0927 1.1239 090 Other low birth weight or preterm ...... 974 886 1.0993 0.0135 1.2 1.0729 1.1258 091 Disorders related to long gestation and high birth weight ...... * * * * * * * 092 Birth trauma...... 5 113 0.0442 0.0197 44.5 0.0056 0.0829 093 Intrauterine hypoxia and birth asphyxia ...... 401 277 1.4477 0.0599 4.1 1.3303 1.5650 094 Intrauterine hypoxia ...... 57 63 0.9048 0.1227 13.6 0.6643 1.1452 095 Birth asphyxia ...... 344 214 1.6075 0.0763 4.7 1.4579 1.7571 096 Respiratory distress of newborn ...... 917 894 1.0257 0.0131 1.3 1.0001 1.0513 097 Other respiratory conditions originating in the perinatal period ..... 1,160 1,372 0.8455 0.0216 2.6 0.8032 0.8878 098 Congenital pneumonia ...... 57 15 3.8000 0.9004 23.7 2.0352 5.5648 099 Neonatal aspiration syndromes ...... 78 56 1.3929 0.1115 8.0 1.1743 1.6114 100 Interstitial emphysema and related conditions originating in the perinatal period...... 146 121 1.2066 0.0595 4.9 1.0899 1.3233 101 Pulmonary hemorrhage originating in the perinatal period ...... 212 145 1.4621 0.0751 5.1 1.3150 1.6092 102 Chronic respiratory disease originating in the perinatal period .... 243 214 1.1355 0.0327 2.9 1.0715 1.1995 103 Atelectasis ...... 382 185 2.0649 0.1144 5.5 1.8406 2.2891 104 All other respiratory conditions originating in the perinatal period . . 42 636 0.0660 0.0101 15.2 0.0463 0.0858 105 Infections specific to the perinatal period...... 563 552 1.0199 0.0261 2.6 0.9688 1.0710 106 Bacterial sepsis of newborn ...... 470 514 0.9144 0.0272 3.0 0.8611 0.9677 107 Omphalitis of newborn with or without mild hemorrhage ...... * * * * * * * 108 All other infections specific to the perinatal period ...... 93 38 2.4474 0.3705 15.1 1.7211 3.1736 109 Hemorrhagic and hematological disorders of newborn ...... 390 274 1.4234 0.0640 4.5 1.2979 1.5488 110 Neonatal hemorrhage...... 319 222 1.4369 0.0698 4.9 1.3002 1.5737 111 Hemorrhagic disease of newborn ...... * * * * * * * 112 Hemolytic disease of newborn due to isoimmunization and other perinatal jaundice ...... * * * * * * * 113 Hematological disorders ...... * * * * * * * 114 Syndrome of infant of a diabetic mother and neonatal diabetes mellitus ...... * * * * * * * 115 Necrotizing enterocolitis of newborn...... 249 203 1.2266 0.0456 3.7 1.1371 1.3161 116 Hydrops fetalis not due to hemolytic disease ...... 120 120 1.0000 0.0264 2.6 0.9483 1.0517 117 Other perinatal conditions...... 1,092 954 1.1447 0.0192 1.7 1.1070 1.1823 118 Congenital malformations, deformations and chromosomal abnormalities ...... 3,400 3,751 0.9064 0.0057 0.6 0.8953 0.9176 See footnotes at end of table. National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 27 Table 2. Estimated comparability ratios for 130 selected causes of infant death—Con. Number of deaths allocated 95 percent according to Estimated Relative confidence limits List comparability Standard standard number Cause of death1 ICD–102 ICD–92 ratio error error Lower Upper

119 Anencephaly and similar malformations...... 299 299 1.0000 0.0 0.0 1.0000 1.0000 120 Congenital hydrocephalus...... 62 91 0.6813 0.0552 8.1 0.5732 0.7895 121 Spina bifida...... 24 32 0.7500 0.0765 10.2 0.6000 0.9000 122 Other congenital malformations of nervous system ...... 191 177 1.0791 0.0477 4.4 0.9856 1.1725 123 Congenital malformations of heart ...... 1,022 1,027 0.9951 0.0081 0.8 0.9793 1.0109 124 Other congenital malformations of circulatory system...... 75 121 0.6198 0.0504 8.1 0.5210 0.7186 125 Congenital malformations of respiratory system ...... 361 571 0.6322 0.0225 3.6 0.5882 0.6762 126 Congenital malformations of digestive system ...... * * * * * * * 127 Congenital malformations of genitourinary system ...... 216 229 0.9432 0.0244 2.6 0.8955 0.9910 128 Congenital malformations and deformations of musculoskeletal system, limbs and integument ...... 269 311 0.8650 0.0319 3.7 0.8024 0.9275 129 Down’s syndrome ...... 57 58 0.9828 0.0705 7.2 0.8446 1.1209 130 Edward’s syndrome ...... 277 278 0.9964 0.0080 0.8 0.9807 1.0121 131 Patau’s syndrome ...... 170 173 0.9827 0.0099 1.0 0.9632 1.0021 132 Other congenital malformations and deformations...... 304 312 0.9744 0.0210 2.2 0.9332 1.0155 133 Other chromosomal abnormalities, not elsewhere classified...... 57 53 1.0755 0.0783 7.3 0.9221 1.2289 134 Symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified...... 2,799 2,732 1.0245 0.0042 0.4 1.0163 1.0327 135 Sudden infant death syndrome...... 2,575 2,485 1.0362 0.0040 0.4 1.0284 1.0440 136 Other symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified...... 224 247 0.9069 0.0270 3.0 0.8540 0.9598 137 All other diseases...... * * * * * * * 138 External causes of mortality ...... 441 444 0.9932 0.0098 1.0 0.9741 1.0124 139 Accidents (unintentional injuries) ...... 292 285 1.0246 0.0107 1.0 1.0037 1.0454 140 Transport accidents ...... 99 108 0.9167 0.0294 3.2 0.8590 0.9743 141 Motor vehicle accidents ...... 85 98 0.8673 0.0368 4.2 0.7952 0.9395 142 Other and unspecified transport accidents ...... * * * * * * * 143 Falls ...... * * * * * * * 144 Accidental discharge of firearms ...... * * * * * * * 145 Accidental drowning and submersion...... * * * * * * * 146 Accidental suffocation and strangulation in bed...... * * * * * * * 147 Other accidental suffocation and strangulation ...... 79 69 1.1449 0.0537 4.7 1.0396 1.2502 148 Accidental inhalation and ingestion of food or other objects causing obstruction of respiratory tract ...... 32 29 1.1034 0.0810 7.3 0.9447 1.2622 149 Accidents caused by exposure to smoke, fire and flames ...... * * * * * * * 150 Accidental poisoning and exposure to noxious substances ...... * * * * * * * 151 Other and unspecified accidents ...... * * * * * * * 152 Assault (homicide) ...... 146 154 0.9481 0.0179 1.9 0.9130 0.9831 153 Assault (homicide) by hanging, strangulation and suffocation .... * * * * * * * 154 Assault (homicide) by discharge of firearms...... * * * * * * * 155 Neglect, abandonment and other maltreatment syndromes ...... * * * * * * * 156 Assault (homicide) by other and unspecified means ...... 91 88 1.0341 0.0417 4.0 0.9524 1.1158 157 Complications of medical and surgical care ...... * * * * * * * 158 Other external causes ...... * * * * * * *

* Figure does not meet standards of reliability or precision; see Technical notes. – Quantity zero. 1Based on the Ninth and Tenth Revision categories shown in table C. 2ICD–10 is Classiﬁcation of Diseases, Tenth Revision, and ICD–9isInternational Classiﬁcation of Diseases, Ninth Revision. 28 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001

Technical notes + Manner of death (coded ‘‘Natural’’, ‘‘Accident’’, ‘‘Suicide’’, ‘‘Homi- cide’’, ‘‘Pending investigation’’,or‘‘Could not be determined’’—see Sources of mortality data figure 1, item 29) on the ICD–10 part of the record was incon- sistent with the cause of death classified by ICD–9. Mortality data used for this report are based on death certificates + ICD–10 cause of death was classified as Other ill-defined and filed in the 50 States and the District of Columbia. The National unspecified causes of mortality (R99) and the ICD–9 cause of Center for Health Statistics (NCHS) receives this information through death did not have the comparable ICD–9 code (799.9). the Vital Statistics Cooperative Program (VSCP) and from copies of + ICD–10 cause of death was an external event of undetermined the original certificates received by NCHS from the State vital intent and ICD–9 cause of death was something else. registration offices. In 1996, all the States and the District of These records were then removed from the comparability file Columbia participated in this program and submitted part or all of the and reallocated to State-specific data files with the intent of mortality data for 1996 in electronic data files to NCHS. requesting the amended information from the States. Comparison of selection and modification rules Coding procedures Table I shows a comparison of the ICD–9 and ICD–10 rules for The underlying cause of death for each record in the 1996 selecting the underlying cause of death. The Ninth Revision rules for mortality data was originally classified by ICD–9 (10). Since 1968, selecting the underlying cause of death are presented and discussed NCHS has coded the underlying cause of death using a computer- in detail in volume I of the ICD–9 manual (9). The Tenth Revision ized software program called Automated Classification of Medical rules are presented in volume 2 of the ICD–10 manual (6). Entities (ACME). Multiple cause-of-death codes serve as inputs to Reliability of comparability ratios ACME, which uses the ICD rules to select the underlying cause of death (13). The ACME system, based on ICD–9 coding rules, was Comparability ratios and other comparability data were pre- used to select the underlying cause of death for all 1996 death sented only for causes of death for which the data were deemed certificates. In addition, NCHS has developed two computer systems reliable. Data not deemed reliable were replaced by asterisks in that provide inputs to ACME. Beginning with 1990 data, the Mortality tables 1 and 2. The criteria for reliability were based on potential Medical Indexing, Classification, and Retrieval (MICAR) system was sample bias and on a close examination of how deaths classified by introduced to automate coding multiple causes of death (12). This one revision were distributed according to the other. The following system assigns entity reference numbers (ERN’s) corresponding to were considered in determining whether comparability data by cause the medical entities listed on the death certificate to specific ICD should not be presented: codes, which can be read by ACME. In 1993, SuperMICAR, an + The absolute number of deaths on which the comparability ratio enhancement of the MICAR system, was introduced. SuperMICAR was based. Generally, if the numerator or denominator of the permits total literal entry of the multiple cause-of-death text as comparability ratio for a particular cause had less than 20 deaths, reported by the certifier, and then converts these literal entries to the ratio was not presented. Data based on small numbers are ERN’s. This information is then automatically coded by the MICAR more likely to be subject to sampling bias. Two important excep- and ACME systems. Records that are rejected, i.e., those that cannot tions are notable in table 2: Diarrhea and gastroenteritis of infec- be automatically processed by MICAR or SuperMICAR, are manually tious origin and Birth trauma. The comparability ratios for both are multiple-cause coded and then further processed by ACME. The based on very small numerators (0 and 5, respectively) and NCHS automated systems have become a de facto international relatively large denominators (144 and 113, respectively). These standard for automating mortality medical coding (18). data were retained to illustrate the dramatic changes occurring for A major task of the comparability study was to reclassify 1996 these causes of death. mortality records by ICD–10. These records were reprocessed using + The percent of deaths by cause from the 1996 mortality data file the ICD–10 versions of the mortality medical software. MICAR-input were included in the comparability file. If less than 50 percent of files for each State used for processing the original ICD–9 data were the deaths for a particular cause (classified by ICD–9) from the used as input to the ICD–10 version of MICAR. The resulting output 1996 mortality file were included in the comparability file, com- was divided into three files for each State: a) records successfully parability data were not presented. Data for causes of death with coded by MICAR, b) records rejected by MICAR, and c) records in disproportionately high rejection rates in the mortality medical which the manner of death was originally pending investigation, but not software are more likely to be subject to sampling bias and are, updated in the MICAR-input files. The records successfully coded by therefore, excluded. MICAR were then processed by ACME to obtain an ICD–10 underlying + An examination of how deaths classified by ICD–10 were dis- cause of death. The resulting ICD–10-underlying-cause codes were tributed according to ICD–9 and vice versa. This showed the then matched to the 1996 U.S. mortality data file by State and certificate source of increases or decreases due to the revision. Data were number to create the comparability file. Records without an ICD–10 not presented for causes of death for which there were large-scale code were then deleted from the file and sent as rejects for manual erroneous shifts of deaths from or to that particular cause. coding. The comparability data were then analyzed to identify likely Standard error of comparability ratios records for which the cause of death had been amended in the final Standard errors of the comparability ratios are calculated using PROC 1996 mortality file. The following criteria was used to designate records RATIO, a procedure in the SUDAAN statistical software designed to as amended: calculate variance estimates for the ratio of two sample means or National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 29 Table I. Comparison of ICD–9 and ICD–10 rules for selection of underlying cause for mortality tabulations ICD–91 ICD–101

Selection Rules

General Rule: Select the condition entered on the lowest-used line of Part I unless General Principle: When more than one condition is entered on the certiﬁcate, it is highly improbable that this condition could have given rise to all of the select the condition entered alone on the lowest used line of part I only if it could conditions above it. have given rise to all of the conditions entered above it.

Rule 1: If there is a reported sequence terminating in the condition first entered on Rule 1: If the General Principle does not apply and there is a reported sequence the certificate, select the underlying cause of this sequence. If there are more than terminating in the condition first entered on the certificate, select the originating one such sequence, select the underlying cause of the first-mentioned sequence cause of this sequence. If there is more than one sequence terminating in the condition mentioned first, select the originating cause of the first-mentioned sequence.

Rule 2: If there is no reported sequence terminating in the condition first entered Rule 2: If there is no reported sequence terminating in the condition first entered on the certificate, select the first-mentioned condition. on the certificate, select this first-mentioned condition.

Rule 3: If the condition selected by the General Rule or Rules 1 or 2 can be Rule 3: If the condition selected by the General Principle or by Rule 1 or Rule 2 is considered a direct sequel of another reported condition, whether in Part I or Part obviously a direct consequence of another reported condition, whether in Part I or II, select this primary condition. If there are two or more such primary conditions, Part II, select this primary condition. select the ﬁrst-mentioned of these.

Modiﬁcation Rules

Rule 4, Senility: Where the selected underlying cause is classifiable to 797 Rule A, Senility and other ill-defined conditions: Where the selected cause is ill- (senility) and a condition classifiable elsewhere than to 780–799 is reported on the defined and a condition classified elsewhere is reported on the certificate, re-select certificate, re-select the underlying cause as if the senility had not been reported, the cause of death as if the ill-defined condition had not been reported, except to except to take account of the senility if it modifies the coding. take account of that condition if it modifies the coding. The following conditions are regarded as ill-defined: I46.9 (Cardiac arrest, unspecified); I95.9 (Hypotension, Rule 5, Ill-defined conditions: Where the selected underlying cause is classifiable unspecified); I99 (Other and unspecified disorders of circulatory system); J96.0 to 780–796, 798–799 (the ill-defined conditions), and a condition classifiable (Acute respiratory failure); J96.9 (Respiratory failure, unspecified); P28.5 elsewhere than to 780–799 is reported on the certificate, re-select the underlying (Respiratory failure of newborn); R00–R94 or R96–R99 (Symptoms, signs and cause as if the ill-defined condition had not been reported, except to take account abnormal clinical and laboratory findings, not elsewhere classified). [Note that R95 of the ill-defined condition if it modifies the coding. (Sudden infant death) is not regarded as ill-defined.]

Rule 6, Trivial conditions: Where the selected underlying cause is a trivial Rule B, Trivial conditions: Where the selected cause is a trivial condition unlikely condition unlikely to cause death, proceed as follows: (a) if the death was the to cause death and a more serious condition (any condition except an ill-deﬁned or result of an adverse reaction to treatment of the trivial condition, select the adverse another trivial condition) is reported, re-select the underlying cause as if the trivial reaction; (b) if the trivial condition is not reported as the cause of a more serious condition had not been reported. If the death was the result of an adverse reaction condition, and a more serious unrelated condition is reported on the death to treatment of the trivial condition, select the adverse reaction. certiﬁcate, re-select the underlying cause as if the trivial condition had not been reported.

Rule 7, Linkage: Where the selected underlying cause is linked by a provision in Rule C, Linkage: Where the selected cause is linked by a provision in the the classification or in the Notes for use in primary medical coding on pages classification in the Notes for use in underlying cause mortality coding with one or 713–721 with one or more of the other conditions on the certificate, code the more of the other conditions on the certificate, code the combination. combination. Where the linkage provision is only for the combination of one condition specified Where the linkage provision is only for the combination of one condition specified as due to another, code the combination only when the correct causal relationship as due to another, code the combination only when the correct causal relationship is stated or can be inferred from application of the selection rules. is stated or can be inferred from application of the selection rules. Where a conflict in linkages occurs, link with the condition that would have been Where a conflict in linkages occurs, link with the condition that would have been selected if the cause initially selected had not been reported. Apply any further selected if the cause initially selected had not been reported. Apply any further linkage that is applicable. linkage that is applicable.

Rule 8, Specificity: Where the selected underlying cause describes a condition in Rule D, Specificity: Where the selected cause describes a condition in general general terms and a term which provides more precise information about the site terms and a term that provides more precise information about the site or nature of or nature of this condition is reported on the certificate, prefer the more informative this condition is reported on the certificate, prefer the more informative term. This term. This rule will often apply when the general term can be regarded as an rule will often apply when the general term becomes an adjective, qualifying the adjective qualifying the more precise term. more precise term.

Rule 9, Early and late stages of disease: Where the selected cause is an early Rule E, Early and late stages of disease: Where the selected cause is an early stage of a disease and a more advanced stage of the same disease is reported on stage of a disease and a more advanced stage of the same disease is reported on the certificate, code to the more advanced stage. This rule does not apply to a the certificate, code to the more advanced stage. This rule does not apply to a ‘‘chronic’’ form reported as due to an ‘‘acute’’ form unless the classification gives ‘‘chronic’’ form reported as due to an ‘‘acute’’ form unless the classification gives special instructions to that effect. special instructions to that effect.

See footnotes at end of table. 30 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 Table I. Comparison of ICD–9 and ICD–10 rules for selection of underlying cause for mortality tabulations—Con. ICD–91 ICD–101

Rule 10, Late effects: Where the selected underlying cause is an early form of a Rule F, Sequelae: Where the selected cause is an early form of a condition for condition for which the classiﬁcation provides a separate late effects category and which the classiﬁcation provides a separate ‘‘Sequela of...’’ category, and there is there is evidence that death occurred from residual effects of this condition rather evidence that death occurred from residual effects of this condition rather than from than in its active phase, code to the appropriate late effects category. The following those of its active phase, code to the appropriate ‘‘Sequela of...’’ category. late effects categories, including those in the Supplementary E codes, have been ‘‘Sequela of...’’ categories are as follows: B90–B94, E64–E68, G09, I69, O97 and provided: 137, 138, 139, 268.1, 326, 438, 905–909, E929, E959, E969, E977, Y85–Y89. E989, and E999.

Rule 11, Old pneumonia, influenza, and maternal conditions: Where the No corresponding rule selected underlying cause is pneumonia or influenza (480–487) and there is evidence that the date of onset was one year or more prior to death or a resultant chronic condition is reported, re-select the underlying cause as if the pneumonia or influenza had not been reported. Where the selected underlying cause is a maternal cause (630–676) and there is evidence that death occurred more than 42 days after termination of pregnancy or a resultant chronic condition is reported, reselect the underlying cause as if the maternal cause had not been reported. Take into account the pneumonia, influenza or maternal condition if it modifies the coding.

Rule 12, Errors and accidents in medical care: Where the selected underlying No corresponding rule cause was subject to medical care and the reported sequence in Part I indicates explicitly that the death was the result of an error or accident occurring during medical care (conditions classiﬁable to categories E850–E858, E870–E876), regard the sequence of events leading to death as starting at the point at which the error or accident occurred. This does not apply to attempts at resuscitation.

1ICD–9isInternational Classification of Diseases, Ninth Revision, and ICD–10 is International Classification of Diseases, Tenth Revision. National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001 31 sample totals (14). Variance estimation for the comparability ratios is significant. Incorporating the variability of the comparability ratio, based on a first-order Taylor series approximation of the deviations of R CM – R estimates from their expected values (19). The general approach z = i,ICD –9 i,ICD –10 taken by PROC RATIO is to form the Taylor series linearization for RSE(R ) 2 2 the ratio estimate. Once the linearized values are calculated, they are 2 2 i,ICD –9 RSE(Ci ) Ci R i,ICD –9 {S D + S D c substituted into an appropriate formula for computing the variance. A 100 100 detailed discussion of the underlying principles used in Taylor series 2 2 RSE(R i,ICD –9) RSE(R i,ICD –10) variance estimation and detailed formulas used by PROC RATIO for √1+ G + R 2 F S 100 D } i,ICD –10S 100 D these calculations can be found in Statistical Methods and Math- The interrevision statistical methods that have been developed ematical Algorithms Used in SUDAAN (19). so far and that are included in Part II of the Guide to State Implementation of ICD–10 for Mortality consider analyses involving Statistical analysis using comparability ratios both crude and age-adjusted death rates, and also cover cases To determine if the change in death rates between 1998 and involving death rates that are based on small numbers, i.e., less than 1999 is statistically significant, the statistical analysis is similar to that 100 deaths. used within the same revision (20,21). Statistical methods dealing with interrevision analyses, however, explicitly incorporate comparability ratios in their procedures because comparability ratios are also subject to random variability. The cause-specific comparability ratio

(Ci) will be applied to rates computed using data from an initial year based on ICD–9 (Ri,ICD–9) to adjust for the quantitative shift that has taken place due to the introduction of the more recent Tenth Revision. CM This resulting value,Ri,ICD –9 = Ci c Ri,ICD –9, is known from here on as a ‘‘comparability- modified’’ death rate. The interrevision methods are discussed in detail in the NCHS publication A Guide to State Implementation of ICD–10 for Mortality, Part II: Applying Compara- bility Ratios (22) available on the Internet at: http://www.cdc.gov/ nchs/about/major/dvs/icd10des.htm. The formulas and procedures discussed in this section are highlights from this publication. To be consistent with previous NCHS publications, the variability of most statistics is described in terms of the relative standard error (RSE). This is merely the ratio of the standard error of a point estimate to the point estimate itself, expressed as a percent. Thus: S(θ∫) RSE(θ∫) = 100 θ∫ where Theta (θ∫) stands for any given parameter of interest and S(θ∫) is the standard error of Theta. Relative standard errors for the comparability-modified death rate and for the crude (not comparability-modified) death rate follow, respectively:

CM RSE(R i,ICD –9) = RSE(Ci c R i,ICD –9)

2 2 2 RSE(R i,ICD –9 ) RSE(C ) RSE(R i,ICD –9 ) = 100 + i 1+ ŒS 100 D S 100 D F S 100 D G

1 RSE(R i,ICD–10) = RSE(Di,ICD –10) = 100Œ Di,ICD –10 CM where C is the comparability ratio for cause i, R − is the i i,ICD 9 comparability-modiﬁed death rate for cause i for the ICD–9 period,

Ri,ICD–9 and Ri,ICD–10 are the death rates for cause i for the ICD–9 and ICD–10 periods, respectively, and Di,ICD–10 is the death count for cause i for the ICD–10 period being considered. These formulas for variability are the basis for constructing conﬁdence intervals and executing tests of hypotheses concerning the change of death rates across revisions of the ICD. The Z-test is the basic test used to assess whether the difference between two rates is statistically 32 National Vital Statistics Report, Vol. 49, No. 2, May 18, 2001

Contents

Abstract ...... 1 Chronic liver disease and cirrhosis ..16 Introduction ...... 1 Infectious diseases ...... 16 Major features of ICD–10 and Alzheimer’s disease...... 17 changes in classification ...... 2 In situ neoplasms, benign neoplasms, Changes in international selection and and neoplasms of uncertain or modification rules ...... 2 unknown behavior ...... 17 Comparability ratios ...... 5 External causes of death ...... 17 Data and methods ...... 5 Infant deaths ...... 18 Results ...... 6 Discussion and conclusion...... 19 Diseases of heart ...... 6 Impact of new revision ...... 19 Malignant neoplasms ...... 15 Application of comparability ratios ..19 Cerebrovascular diseases...... 15 References ...... 20 Chronic lower respiratory diseases . 15 List of detailed tables ...... 21 Influenza and pneumonia ...... 15 Technical notes ...... 28 Diabetes mellitus...... 16 Nephritis, nephrotic syndrome and nephrosis ...... 16

To receive this publication regularly, contact the National Center for Health Statistics by calling 301-458-4636. E-mail: [email protected] Internet: www.cdc.gov/nchs/

Suggested citation Copyright information National Center for Health Statistics Anderson RN, Miniño AM, Hoyert DL, Rosenberg All material appearing in this report is in the Director, Edward J. Sondik, Ph.D. HM. Comparability of cause of death between public domain and may be reproduced or Deputy Director, Jack R. Anderson ICD–9 and ICD–10: Preliminary estimates. copied without permission; citation as to National vital statistics reports; vol 49 no. 2. source, however, is appreciated. Division of Vital Statistics Hyattsville, Maryland: National Center for Health Director, Mary Anne Freedman Statistics. 2001.

DEPARTMENT OF MEDIA MAIL HEALTH & HUMAN SERVICES POSTAGE & FEES PAID Centers for Disease Control and Prevention CDC/NCHS National Center for Health Statistics PERMIT NO. G-284 6525 Belcrest Road Hyattsville, Maryland 20782-2003

DHHS Publication No. (PHS) 2001–1120 1-0355 (5/2001)

OFFICIAL BUSINESS PENALTY FOR PRIVATE USE, $300